The phenomenon with VS picking up a tail-on target out to 50km but head on less than 15 sounds more like it's an issue of using EOS rather than the radar, in which case it's correct - not a bug.
I don't know what to say about the radar one. To me it sounds like it would jsut introduce too many false returns and lock onto anything it bloody well pleases, which is why the ranges are limited on these modes, AFAIK.
The MiG-29 radar's history is fraught with clutter problems and inadequate processing power, neither of which, AFAIK, is modelled in LOMAC. Incidentally the Su-27's radar, being essentially the same, should be suffering from the same problems. It's probably pretty difficult to model however, and who the heck knows what the false positive rate was on the Eagle's radar, so may as well ingore the whole thing....