I know, sorry, my english doesn’t help me much either. Let me try in Korean.
If you re bored to read it all, just read the bolt letters:
In a beam that is resting in supports at its both ends (first diagram, called “free standing” in Chinese and “simply supported” in korean) for a uniform load Q, the maximum deflection (in the middle of the span) is 5QL^4/384Ei, where L is the span.
If you extend that beam both ways and place multiple supports at equal distances keeping all spans =L, then the deflection for that same span will be QL^4/384Ei, that is 5 times less than in the previous case. (The load in the neighboring spans act as a counter weight reducing the deflection. You can see in the first diagram, that just beyond the supports, the beam is deflecting upwards. In an arrangement with multiple supports that upward distortion is reducing the deflection in the middle of the neighboring span. I called that “a continuous beam” in Chinese and in korean.
In the past the Chinese have tried to build a continuous wall around the earth, but not a infinite beam around it. Not yet. So all beams, even the “continuous”, have ends. At the end spans of you design the counter weight effect comes only from one side. So there the deflection it will be less than what is in a “simply supported” beam with equal span, but increased (~doubled) compared to the typical middle span of a “continuous” beam.
That’s why in an optimized arrangement of multiple supports, the end spans should be sorter.
In both beams (simply supported & continuous) deflection is proportional to L^4. That means that if the span (L) is doubled the deflection (f) is expected to grow 2^4=16 times.
In your second drawing you have a design with 3 extra pairs of vertical supports. Alex calculated the maximum deflection with out them. By reducing the span with these supports to 1/4rth of what it was, the deflection would be decreased 4^4=256 times. More reduction (>2*256=512) is expected, as the beam is now “continuous” and not just “simply supported at it ends”.
About what I wrote in perfect Chinese for skewing and caging:
I missed that you were talking imperial (god, That is Chinese) and thought that
the tubular 5x5 are 2,5mm thick. My mistake. Since they re
6mm thick, probably welding them together is enough to transfer loads from the the middle SHS to the outer's SHS’s walls where the extra supports will be joined. Still, it might be easier to use cages for the xtra supports, instead of stitch weld the three 5x5 together.
Usually flat bars are cheaper than tubular sections for the same weight. Plus, pushing a pin through a tide-fitting hole is easier than through 2 tide-fitting holes (no laughs pls). So making the extra vertical supports with flat bars instead of making it with hollowed 2x2s might be more convenient.

Hope it made some sense,
fa
edited/ in the kingswood video @~4:50 there s a press with multipl twins flat supports