Splat, BH is right. The primary function of the tail is anti-torque. In fact many full size helis only use the tail rotor to yaw in one direction. Yawing in the other direction is accomplished by the torque of the main rotor with the tail rotor at zero pitch. The MD520N NOTAR you spoke of gets much of it's "anti-torque" from the rotorwash passing over the outside of that large tail boom. This is called the Coanda effect. So the sliding vent in this case acts more like a rudder, but understand the series of McDonnell Douglas 500 helis are more of an exception than a rule. They're basically a fighter jet without wings. They can loop and roll without problems. I've also heard of them referred to as the "Porsche of helicopters".

The rotor doesn't need to spin in the opposite direction for an auto. This may help explain that. Instead of thinking of the rotor head as individual blades think of it as a rotor disc. Then think of the disc as a wing on an airplane. After all it works the same way. This is why forward airspeed is important in real heli autos. Also in real heli autos RPM managment is critical as you can still overspeed the rotor. On RC helis I imagine just get all the headspeed you can.

Autos are basically a controlled fall. You start by dropping the collective. This allows the flow of air coming up through the disc to keep it spinning. The low collective setting has no lift, so it has no load therefore no reason to stop turning. Like a pinwheel in the wind. Keeping RPM's up is critical because the RPM's in your rotorhead are the only power/stored energy you have to arrest your fall. A certain amount of forward airspeed is also necessary because it not only allows more wind through the disc, but dramatically speeds RPM's at the flare. Increasing the collective at point of flare allows the RPM's in the disc to produce lift. Ideally this allows the heli to settle to the ground gently.

This is over simplified quite a bit, but I hope it's enough to get the idea.