EFI is on my roadmap as well. All my previous project cars have been EFI and though I never delved into mapping or programming them, building up a standalone EFI powerplant has long been a dream of mine.
Thus when I got my A body, I went searching and the best walkthrough, so to speak, I found was this:
http://www.dodgecharger.com/forum/index.php/topic,99432.0.html
It's even in a BB Mopar ;)
The DIY route isn't terribly difficult, but does require lots of effort to scrounge the needed parts and if you want to keep it under a tight budget cap most of that scrounging will take place in junkyards rather than online. The fabrication work tends to be relatively minor, the biggest one being injector bungs in the manifold since EFI manifolds tend to leave fewer choices in the aftermarket. Injector placement depends mostly on the convenience of the fuel rail. Closer/farther and angle relative the port don't have huge effects, and the effects they have tend to be RPM dependent. Reliability and a decent idle tends to favor closer to the head and about 45-60 degrees relative the port flow direction with enough protrusion to get the fuel into the air (rather than on the wall of the port) without being a major airflow obstruction. Adding the proper fuel system to the tank is about as much work as the intake, depending on the fabrication tools one has access to.
Sequential fuel injection has benefits, but typically only at lower RPM since the pulse width of the fuel pulse at higher revs will tend to last longer than the valve opening event. As a result, batch fire can be "just as good" as sequential unless max MPG and efficiency is the goal. Sequential typically requires higher flowing injectors as well, since they're (typically) only firing once per valve event vs batch fire. This can have a deleterious effect on idle quality since high flow injectors tend to not flow nicely down low. So the benefits can be quickly outweighed depending on the nature of the engine build. Luckily, it's easier to make a sequential system operate in batch-fire mode than to upgrade a batch-fire system to sequential...
The ignition tends to be the trickier part of the equation, but also isn't rocket surgery. If/when I do my own system I plan to keep the distributor (batch fire with crank trigger only) until I get a handle on things, then expand to ignition. It's mostly a cost and learning curve issue.
As with most things, how much capability you need depends on what you intend to do with it. An EFI system for a mostly strip oriented car will be very different from a grocery getter. Wanting to map and program the ignition vs keeping a distributor is also dependent on goals. Knock sensors can be utilized to get the most out of a timing map, and the programmable ignition makes it easier to compensate for crappy gas in a roadtrip type car.
