The porting of a log head is set up for a 30 to 40 degree draft angle. The first runner 1 into number one cylinder is like 20 degrees from horizontal. The last runner number 6 into cylinder number six is like 40 degrees from horizontal.
Sooo, if you angle mill the two faces of the log (with a 12 to 24" mill), and do so at about 30 degrees to the horizontal, two CFI units will do a great job feeding 3 cylinders each. And you can keep the carb flange and exhast flange as it is, and still clear the spring towers.
A few points.
1. The particalar Ford or Chev or Mopar CFI units which are 1 or 2-bbl imitators are inferior to most carbs in fuel atomisation. It's a fact noted byThe Frenchtown Flyer, and by various anectodatal info from TBI units used on 1982 to 1993 ish Chevies, Chryslers and Fords.
2. There are some manifolds which suffer compound problems when a CFI unit is added. They are, specifically, the swan neck manifolds with sharp 90 degree radii from the carb to the manifold entry point. Even worse are those which have a technically greater than 90 degree radius from the carb to the intake valve. I'm very familair with the Lynx style Twin DCOE type manifold used on that 429 Landau. Its a great performance carb set-up, but it is the worst part throttle or economy set-up around. It was used variously by XA, XB Falcon GT Hardtops from 1973 to 1977, and again on the TR7 V8 rallye cars campaigned by Leyland in 1978. They were awfull for fuel economy.
Rule is that if you reduce the short turn radius, keep the bends to a minimum. Doing a minium bend will improve on the bad CFI fuel atomisation, and stop fuel drop-out. I'd err on the side of caution. If there is evidence of fuel leaks on any 90 degree CFI set-up, then I'd never, ever put it anywhere near the exhast pipes.
3. Funny thing is a down draft CFI with the right 90 degree bend on a good manifold will work just fine becasue the mixture motion is mildly supercharged by the free fall, even if its only 2" or so.