Unitary Machinery Support was a Woodard idea from the late '20s that put 'larger' cylinders close together in an attempt to reduce augment. It's more noticeable on the 2-12-6 (there was a plan view in Woodard's presentation to a fuel-related organization in 1928, which is where I first saw the drawings).

This was a rational approach to reducing the amount of overbalance that would otherwise be necessary to compensate for nosing (and coupling into hunting) and hence the effective rotating imbalance that causes hammer blow at speed. In my opinion the work on better balance with lightweight reciprocating parts, which essentially started with Eksergian (also in 1928) is the line of practice that 'won out' -- rather compellingly when you see the effect of its implementation on designs like the T&P 600s, which were immediately-pre-UMS Woodard in virtually every detail when built.

I have a suspicion that this "Hudson" is intended to be a 3/4-size 2-8-4 (for railroads that want the radiant advantages of a large firebox but don't need eight-coupled power) but with better leading-truck action that may be required for yaw effects on the shorter coupled wheelbase. This perhaps more than anything 'dates' the design; it should be near-obvious that the length or other advantages of this design over a 2-8-4 of otherwise comparable dimensions are minimal, the gains in adhesion both at starting and at speed from an 'added' driver pair considerable, and (as AMC would increasingly prove in very short order) there is little reason a good 2-wheel lead truck design can't make the speed required of any freight engine, and much effective passenger service as well...

I have seen references that attribute the articulated trailing truck to American Arch (or all people!) as the actual patent holder (I think the link provably involves LeGrand Parish, who was head of American Arch and became president of the 'new improved' Lima in mid-1918...). It is a classic 'nifty design' that doesn't hold up well at all when required to do things not expected in the original design ... like back up a train at high cylinder horsepower. Or through switches and crossovers...

Not to mention that it provides laughable control of the rear end of the abbreviated chassis when it comes time to counteract the effect of ... piston-thrust nosing. There are many good reasons why four-wheel trailers came to be Delta-type with some of the equalized weight-bearing being linked close to the pivot, but most of the bearing imposed as far back, and laterally, as available clearances will permit, to give the best steering effect at the rear.

Over the years, you see repeated experiments with pushing the front tender truck up 'as far as possible' and using better linkage or buffers to have this truck actively 'control' the rear of the locomotive chassis better. This approach is virtually essential on a high-speed 4-6-0 or 4-8-0 where stiff lateral compliance from the lead truck is insufficient or zero overbalance design practice can't be observed.

Something I don't know is whether Lima provided the cuts used in the article cited, or whether a separate artist, perhaps from the publication, did the actual version printed. I'd suspect, but can't prove, that Woodard or some responsible person at Lima either provided drawn views (they are not 'diagrams' in the usual railroad sense, but side elevations) originally, and those were then adapted as needed for print.