I'm not saying you couldn't make a wood version. That was the original intent with the P-39 that I made a mockup for. I was considering the time required. BJC mentioned the GP-4. That's a great aircraft that takes 5000 hours to build from a set of engineered plans. I don't even have the engineering nor the plans for the B-25 yet.
The intent for the B-25 was to use a steel tube frame with hot wire cut foam bonded to the frame and a composite outer skin like one of Rutan's aircraft. Think of a medical CAT scan with all the slices. Foam billets are 10 x 20 x 96 inches making each of the "slices" 10 inches thick. That makes for a lot of foam blocks, but if a cnc hot wire cutter was used it would be manageable. Working from a solid model of the aircraft and inner steel frame, the foam blocks would fit onto the frame tubes and be bonded to them. A minimal amount of sanding would be required to fair in the seams to prep for the outer skin of glass or carbon fiber and epoxy. The steel frame would handle all the point loads and provide the distribution pf loads throughout the aircraft. Possibly the outer wing panels and horizontal and vertical tails would be conventional aluminum construction as in an RV. The design and build of these structures are straight forward and there's lots of RV builders where I live that I might recruit.
It would be nice to build the entire B-25 in conventional aluminum fashion, but we have been trying to come up with a construction method that would take fewer manhours. I'm thinking of the Bally B-17 as an example. That project is an all conventional aluminum structure and took over 10 years and somewhere between 10,000 and 20,000 manhours to design and build. Bally did a tremendous job in bringing his dream to completion, but it took so long he didn't have much time to enjoy it.
I sat in the mockup last night and there's enough room in all directions. The forward visibility is similar to that in a Cessna 195. Photos in a few days.
If you make your 4130 frame relatively square in the bulkhead sections and put tabs or sheet metal screw strips along the longerons you can use the foam for a mold and just attach the shell along the screw strips. Non-structural skin in other words. You use Boku's method with the pool noodle stiffeners. I am guessing almost all the surfaces are flat wrapped or close to it. That would be quick and easy. I suppose you could do the same with a few formers and aluminum skin for that scale look. There are a lot of hybrid steel tube and wood, fabric, composite, aluminum fuselage structures out there.
The B-25 mock-up is progressing well. I have attached a photo as a teaser.
Obviously, the construction is nothing like any method one would use to make a flyable version. I use a lot of double wall corrugated, and 1/4" & 1/2" shop grade plywood. The ME-262 was built the same way. I have sat in it and there's plenty of room in the pilot's seat. The passenger seat had to be moved aft several inches. As a result a 24" extension was added to the fuselage along with a short section of the wing "D" section leading edge.
In a 50% version you don't have to put the passenger in the turret. Unfortunately, in any scale less than 50% the pilot has to be reclined or the fuselage made taller. The P-61 would make a very good scale project. The only big downside to these twin projects is the cost of the engines and props. Verners start at $16K and Rotecs about $18K. Since you don't want a windmilling dead engine the props have to be in-flight adjustable bringing the engine/prop bill to at least $50k.