[mati140]

1. Maybe you can map the shift between automatic - manual only mod to one of difficulty level options? For example "limited ammo" or "no padlock" ?

2:

While that's true for later versions of the AIM-9 for sure, I watched a set of videos showing launches of early stage AIM-9 and AIM-7 missiles, and those which have been mounted underneath wings did drop that way for real.
There's a significant difference to later version AIM-9s mounted e.g. on wingtip stations. Those missile's had their rocket engaged prior to releasing them from the rail, thus they "swooooshed" off the rail. In contrast, the very early guided missiles didn't seem to have rocket motors reliable enough for that task, hence the drop to stay clean from A/C in case of rocket motor failure.
However you're right for any later version of the AIM-9. An AIM-9D for instance couldn't leave an F-8 with such kind of drop pattern, it has to have it's rocket engaged on the rail.

Are you sure it did deatach like this though it was mounted on rail? Maybe it didn't use rail but pylon? And maybe after modelling the few seconds arming delay after launching missile maybe there will be no risk of being blown up by own missile?

3. I'm afraid that one of things that are uber for early version is the missiles FOV. You said it has 4* seekers FOV but with mirror the FOV is 25* (or 30*, I don't remember). However, althuogh the whole field that mirror can observe is 25*, the angle that it can see at one time is only 4* that seeker looks in at the moment - and at the begining it looks straight. It won't even know about the enemy inside 25* FOV untill it pass trough the central 4*, gets spoted by seeker and starts being tracked - the lock want be broken as long as enemy stays inside 25* FOV then. Am I wrong?