Well I simply brought up alternating, because it's the closest you can get to simultaneous, without running the battles in some sort of real time engine. In lieu of real time running, there are serious snags with any potential setup - including what the game currently has - but at least what we have now doesn't require a significant change of game mechanics, AND can be mitigated somewhat by slowing research so as not to get all magic schools at 9 by turn 50-60, which is the only reason that this issue should ever come up. With slower research, forcing choices in spells means that even in the "late game" you are struggling more often with checks and balances, rather than playing a game of "lets all do everything all at once until we get sick of it or someone just wins, possibly because the other guy staled from being sick of it".
Ultimately though, I think that perhaps the most elegant solution to this actual problem, is to analyze the spell itself. Perhaps the best way to avoid this problem, is to change Master Enslave from affecting the entire battlefield, to simply having a huge AOE, such as the massive endgame evocations. In this way, one person having the spell is still a huge boon for his war effort, but both sides having it leaves much more room for an actual magical tug of war for supremacy, and it makes multiple casts of the spell into something that is much more vital to its effective use.