In addition to all the other reasons:

 - Every camera-to-screen system has lag. Lag in a dogfight can be fatal.
 - There will be some distortion between the camera and the screen, no matter how well designed. Distortion can be fatal too.
 - Cameras and screen cannot reproduce depth perception (yes there are ways to do it but they are a LOT more complicated than cameras and screens). Lack of depth perception can be fatal in lots of situations.
 - It is possible to get positional cues (especially depth cures) for a visible object by moving your head, even the small amount of movement that a fighter pilot's head can make. A camera screen system won't do that.

A screen system can be useful if the alternative is not being able to see in some direction at all, but you wouldn't replace direct vision with it.