Plans for Partial Alternator Elimination

I've been thinking about this for a couple of days, and after reading about mpg gains in the 10% range on ecomodder.com via removal of the alternator belt, reading about GM's EV-1 plug-in electric car (which worked fine, but they destroyed because... well who knows) and the new, exciting aptera, I've decided to hybridize the Gascort.
Removing the alternator belt is good for testing, but not especially valuable for long-term use. In addition, the computer, fuel pump, and injectors rely on a 14-15V source from the alternator; the battery only runs about 12-12.5V at best, so running them on the batteries when they were designed with the alternator in mind can actually cause worse performance of injectors and sensors, hurting fuel economy. Several other ecomodders have done this with no ill effects and increased FE, but I'd like to modify their ideas a bit for my own use. I'm planning to lessen the load on the alternator by only having it attached to crucial electronics and disconnecting it from the rest of the car. This should work to increase horsepower and reduce drag, because alternators vary the amount of resistance inside them to correspond with electrical demand. When my buddy Brandon had a huge (2000W at least) stereo in his car, you could feel the car dragging hard when he turned on his amplifier.
Whoa..Wait a second there, professor. You can't just disconnect everything else and have it magically charge up by itself, for free! That energy has to come from somewhere, doesn't it?
Yes. I'm planning on plugging my car in at night to charge for the non-integral systems, namely the lights, radio, wipers, blower fan, and the starter. Eventually I'd like to add a solar panel to charge while the car's parked at work.
How does this fit in with the whole saving fuel bit? Here are my three arguments as to why I'm doing it:
First, consider that you're using a relatively cheaply made alternator to take mechanical energy and turn it into electrical energy. The electric utilities have much larger and thus more efficient units with less mechanical energy wasted as friction.
Second, consider your source of mechanical energy. Your car, with its fairly advanced emission controls and fuel-injection technology is good at extracting energy from gasoline, but isn't as optimal as it could be.
Third, consider the cost of fuels from which we're making electricity. Coal (our traditional source) is WAY cheaper than oil and will continue to be for the foreseeable future; we have at least 300 years of it left. Nuclear, hydroelectric, wind, and solar are cheaper as well, and will only become better (cheaper than the equivalent amount of energy from oil) in the future.
Google.org (Google's philanthropic group) is developing plug-in hybrid technology now with their fleet. Check it out - it's pretty cool and it drives the effective efficiency of their hybrids up considerably. They consider their gas MPG and log the cost of electricity used to charge them overnight for a cumulative MPG total.
http://www.google.org/recharge/

Here's a schematic of my plan. (click to enlarge) It allows for a cheap lawn mower battery to be used as a capacitor in the alternator - ecu circuit, and a source of power as the ecu needs to be on before the engine is turned over.
The regular car battery remains in its current state, wired to everything else on the car. I'll install a heavy gauge wire and a heavy duty "charge switch" between the two systems so I have the option of using the alternator as it's typically used. I think I'll try to use it when decelerating at offramps and down hills as well. The reason I want it to be capable of 50-60A is so if the big battery is drained down past the point where it'll turn the starter, I can use the lawn mower battery to do the job (infrequently, for sure, but a nice emergency jump-starter).
I will need to get a volt-meter to monitor my big battery's charge so I can charge it before it goes dead. For now, I'll probably wire my Digital MultiMeter (DMM) in to watch it.