LostSoul said:
I think this is real problem with GM supervised effects. But if you gave the Players some method of, with thier spells, effecting the plot line or improving the effect of thier spells, it might work ok.
Note what Thorin Stoutfoot said about de-empowering a player. If the players don't have a solid handle on what will happen, they won't like it much. You give the example:
"As well, I'm going to use the 'unforseen benefits' clause so that sometime in the future this will have a positive impact on us that we didn't intend."
This gives the player zero idea of what is actually going to happen. Some unspecified positive impact some unknown time in the future? What magician worth a darn would attempt to operate under such conditions? "Hey, guys, I just cast a spell! Maybe, something good will happen in a week or two!"
In a fantasy novel, mysterious magic is cool. The reader and the characters may not fully understand what's going on, and that adds tension. But playing the game isn't like reading a book. It's more like cooperatively writing a book. And if the author doesn't know what's up when magic is about to happen, then there's a big problem in your book.
As a couple have mentioned, magic in D&D becomes a lot less dry if you just take the effort to be descriptive. If the wizard's player says, "I cast Fireball" nothng special is happening. If he says "A small jet leaps from my hand, and swells into a 20 foot ball of searing mystical flame," things are more entertaining. It isn't the system's fault if the players and DMs aren't descriptive at the gaming table.
Rather than giving the magic system a huge overhaul, try something small to encourage descriptive play. Say the save DC goes up by one if the spell was well described (or give a -1 if it wasn't described, or something similar). Encourage descriptive use of magic, and much of the percieved problem may go away.