Experienced railway guys would probably know which way the handle was for which way the points were set, but the general rule is that they are left on the main route rather than a setting which would take you onto a siding, so in theory at least, you would switch the points back to the main line after you had passed them, but in Railworks it doesn't matter so much if you don't. Note that there are keyboard keys to switch the points (G switches the point in front of your train for example). In the real world, this is why both locomotives and cabooses have steps and platforms at either end, so that the train can move slowly and someone can ride on these platforms or steps, hop off, switch the points and then hop back on, and in the real world, people who do that are trained to grab the railing that is nearest the direction of travel so that they get hauled up onto the steps, because if they grab the rearwards railing, it will swing them under the wheels. This is also one of the reasons why speed limits are slow in yards, so that people can get on and off the loco or caboose in order to operate switch levers. You can simulate doing that in in Railworks with the shift+2 key combination.
Of course as in real life, you will come across both manual point switches and automatic ones that are controlled from a signal box or control centre. In the case of those controlled from a signal box, there is invariably what is known as a 'Block System' in place. This is where the route is divided up into blocks with signals at the start and end of each block. There will be a board up on the wall in a signal box with a diagram of the rail layout and lights on the board in each block. As a train enters a block, a light will go on up on the board in the sector the train has entered to indicate on the board that there is a train occupying that part of the track (these can be triggered in a number of ways, although usually it is via a magnetic system with something under the locomotive that triggers a detector on the track (this usually located right underneath the front end of the loco and invariably hidden from veiw by the couplers and snow plows etc). More sophisticated systems installed on tracks can actually count how many wagons are going over them too and indicate that count to a controller via his signal board. Some track systems can even send a radio signal with an audio recording of a voice message which informs the driver of any potential problems ahead in a manner similar to the recorded hourly information that pilots get from airports when they tune the appropriate airport frequency.
To keep things organised, when one train enters a block, the signals by the trackside change to indicate that block is occupied, and when the train passes into the next block and triggers the next block detector, the signals change again to indicate the previous block is clear. That's what all those cables running alongside the tracks do and why there are sometimes delays on trains, because thieves sometimes steal the wire for its scrap metal content, and occasionally electrocute themselves in the process.
Thus there should only be one train in each block at any one time if things are going to plan, although in switching yards, things are often done simply by looking out for danger, so that block system is more geared towards preventing collisions on main lines. The Block System is pretty much universal on railways around the world, since it is a simple yet effective means of keeping things safe, although in more recent times, it has been augmented with automatic warning and protection systems which trigger an alarm if a train has passed into a dangerous location or is getting near another train, which is why such a system isn't much use in a switching yard, since by their nature, locomotives in switching yards have to trundle up to other wagons in order to couple them up.
In modern signal boxes and control centres, the system is even more sophisticated, whereby instead of simply a light on a board indicating that there is a train occupying a block, there are what is known as 'train describers', which give the signal and switch operators information about the train and where it is going, thus the controllers know which switches to throw in order to route the train correctly. On some trains around the world, this info also used to appear on the front of the train itself on the headcode board numbers, although in modern times, this has given way to computer systems. But if you ever see pictures of, for example, 1960s and 1970s locomotives in the UK, you would see codes such as IE06 on the number boards at the front of the loco, which could tell you what train it was and what route it was on if you knew what all those codes meant. In the US however, the numbers on the front of a loco are simply the locomotive's inventory number, so the route identity code system was by no means universal. Nevertheless, all that block signaling system was mostly well established before the 1920s and even things such as train describers were fairly common by the late 1930s, so this is a system that has been in place for the most part, for 100 years or so, and the system we have today is really just a refinement of that which incorporates computers as opposed to cables, pushrods and telegraphs.
In addition to all that signal and indicator malarkey, a real locomotivce and indeed a locomotive in Railworks too, has a built-in warning system. If a train goes into a sector where there is some sort of potential danger, an audio warning sounds and a button in the loco cab has to be pressed to cancel the warning, which if left unchecked, would then trigger the brakes automatically and stop the train (Q is the key to cancel it in most train sims). This is what the 'dead man's handle' principal which you have probably heard of is all about, i.e. if the train driver dies at the controls and is therefore unable to cancel a warning, the train would stop, although in some locomotives, there is also the additional safeguard of the driver actually having to hold the throttle in position, so that again, if he was incapacitated, his hand would slip off the throttle. In films where they have runaway trains, they usually have to come up with some clever excuse to get around that, for example, in the movie 'Runaway Train', they have the brake blocks burn out and the system fail so that the train can run away when the driver dies. Since that is extremely unlikely to happen in real life, when the Alaska Railroad, which was used to film Runaway Train agreed to let the film-makers use their system to make that movie, they insisted that the locos did not have their railroad name on them, which is why the locos in that movie all look mean and moody in their black/brown paint jobs.
To get a better handle on what all that 'lap' 'self lap' 'released' and 'application' stuff is about on train brakes, I'd recommend having a read of the following link (that website is worth bookmarking by the way):
http://www.railway-technical.com/air-brakes.shtmlWith regard to couplers, it depends on which country you are in (or simulating being in). In the UK, and indeed most of Europe, the buffer and chain system is still very common on freight trains, i.e the buffers separate the cars, but the chain attached to hooks on the cars is what connects them, so that's not a rigid system of connection. In the case of the UK you can expect that to change eventually, since UK freight is in fact often owned by the Burlington Northern and Sante Fe railroad (i.e Direct Rail Services is owned by BNSF). That American ownership has influenced locomotive design in the UK, notably with the Class 59 and Class 66 locos, which are American SD-40-2 locomotives with a UK-style body grafted onto them, so it is also likely to influence stuff such as couplers as well in the end. Until such change happens however, the buffer and chain system will still be common in the UK, and that is indeed cushioned by the buffers being spring loaded. It is also why guards vans in the UK were often referred to as 'brake vans', because in order to keep the tension on buffer and chain couplers manageable, UK freight trains were often run with the brake van having its brakes partially applied (brake vans are also heavily weighted to assist that effect too).
In most of North America, and indeed much of South America too, wagons are coupled using the Janney coupler, although it is almost never actually referred to as a Janney coupler, and is more commonly called a 'knuckle coupler'. This is a far more sophisticated method than the buffer and chain system, which is why you are starting to see a similar type of coupler on most European commuter trains, including in the UK. The knuckle coupler is cushioned to some small extent too, although there are in fact several variations in design, the most modern of which actually has built-in deformable bolts behind it which act as crumple zones in the event of a crash, but in any case it is rigid enough to keep trains under sufficient tension to make buffers unnecessary on US rolling stock, since a knuckle coupler actually weighs about 50 lbs, so it is a pretty tough component by itself, which is a good thing. The knuckle coupler has the advantage of being automatic too, i.e you smack two knuckle couplers together they will hitch automatically, leaving only the brake pipes to be connected manually. This is much safer, since you don't need someone down on the track between the wagons when actually coupling them. The disadvantage of the knuckle coupler is that because it is more rigid than the buffer and chain system, if you put too much tension on the train with acceleration, you can break the couplers and it is not unknown for freight cars to have been torn in half before the coupler actually broke, although it is worth bearing in mind that US freight trains are often over a mile long, so that's a lot of weight and momentum to be dealing with. Eastern European and Russian trains have a different coupling system, called the SA3, but it is basically very similar to the US knuckle coupler.
So for any of those coupling systems, you need to be smooth and take up the slack gently, then keep it under tension as best you can. You then slow down as gently as you can too, so as to avoid the wagons all smacking into one another if operating a UK train with the buffer and chain couplings. One other thing to bear in mind is accelerating on a curved section of track, which can cause a derailment if you do that too violently. To understand why that is so, picture a toy train on a circular track and imagine pulling the loco very quickly, the force will pull in a straight line from the loco to the tail end wagon rather than through all the couplers and all the wagons will tip over off the track. This is known as a stringline derailment, and is one of the reasons why on long trains through curvy tracks (common in mountainous terrain in the US), there is often a loco at the back of the train and one in the middle of the train, so that the force pushing and pulling the train is not all coming from the front end of the train. This is called 'distributed power'.
If you want some useful info on how and when to apply power to a loco, then I can recommend looking here, where you will actually find many real manuals for locomotives:
http://gelwood.railfan.net/manual/manual.htmlHope that helps.
Al