The rules still produce reasonable results for non-nuclear combat. Ignore spillover fire, of course. And that is the only GEV rule that actually reflects nuclear combat. EDIT -- the optional rules allowing the ruining of bridges and town hexes should be ignored as well.
Beyond that, a Heavy Tank will defeat a Light Tank most of the time as in real life. A missile armed tank has a decent chance of hurting the Heavy Tank if the missile tank gets first shot. Since it has double the range, there's a decent chance of this. If the missile tank fails, the heavy will likely destroy him when he gets in range.
As far as infantry, don't worry about it. Give each squad a defense of 1 and the results will be good enough.
If you have armored and unarmored infantry in a battle, give the unarmored infantry a 1/2 defense strength per squad.
One tweak that could be made would be to introduce an overwatch rule. From my rules A Fistful of TOWs 3 -- a unit that is eligible to fire may instead take an overwatch marker. This allows it to fire before, after or at any point in the enemy movement phase. If overwatch fire happens before or after the enemy movement phase, no modifiers apply. If the fire happens during the movement phase, apply a -1 to the attack roll.
A faster rule that gets about the same result would be the FFT Hold Fire rule: a unit that is eligible to fire may instead take a hold fire marker. This allows it to fire in the enemy fire phase. This fire happens before the enemy fires.
Or you can ignore overwatch. The very low speeds make panzerbushing less of a problem than one might expect.
If you are fighting a largely infantry action, the original Squad Leader game and rules can handle that superbly. Use the same approach -- everything is rated relative to what's in that fight.
This approach will produce *rational* outcomes; it will not replace dedicated wargames like A Fistful of TOWs or even Striker. But GEV or even SL are plenty fine for RPGs.
I think the description of a "D" result isn't all that good... because it mentions (at least in some editions) that it's an EMP and shockwave effect; the time down is due to things rebooting as much as it is from the shock of fire nearby.
I'm not an expert on armored combat, tho'
Also, there's the odds issue (called out in The Ogre Book, Playing the Odds in Ogre) which seems a bit of a counterfactual approach for anything short of nukes... Worst case:
2x 1:1 is mathematically far superior to 1x 2:1.
1x 2:1 33% D 50% X or XX
2x 1:1 22% D 67% X or XX
That's fixable by tweaking the table (either adjusting the odds or the result rolls).
I'm also not fond of the odds hiccup between 1:2 and 1:1...
Plus, the all-or-nothing of a hit on a disabled unit outside overrun being an autokill.
I think using the overrun values across the board for disabled, but not auto-recovering might make more sense; turn it into a damage step rather than a condition.
Note: There's an article on using OGRE for 2300 in one of the magazines.