I've found in many of my games (allowing for the currently minimal AI) that with many of the tactical spells, whoever casts first gets a huge advantage.  Late game it turns into beat-with-a-bigger-stick, and this is one reason why fodder units are so useless at times (fireball + a few fire shards = armageddon for any army)

So here's an idea for a mechanic:

• When a mage casts a tactical spell on the other army, there is a chance that a mage in the other army can counter it
• This can either be blocking it completely (unlikely) or reducing its power (more likely)
• How successful the defending mage is is tied to the magic levels and class of the mages
• If the defending mage is adept at the class of magic of the spell being cast, he or she will have a good chance of doing something about it
• level 4 can still diminish a level 5 spell, but by no means as much
• There is chance in it, as with regular attacks, but the chance should be heavily weighted by the abilities of the mages involved
• It does not apply to spells on the mage's own army

What do you think?  It would need balancing, and we don't want to make mages useless, but it would certainly bring more of an element of skill in choosing your mage and your spell given who is in the other army.  It would also present a way to stop a high level mage with initiative from nuking everything late on (while still being able to take out fodder spam without good mages).