That's why I lobbied for formulas in Traveller5 (and got them).
Well, I mean, that's all well and good.
But it's not "how things work". I mean, to a point that's true, but I don't know how much a Saturn V rocket is simply an "upscaled" Redstone motor. Is it just a material science/manufacturing issue of making bigger nozzles, pumps, and hoses?
One thing we don't see in Traveller is what we see today. Stacking motors.
The Falcon 9 and Falcon Heavy use the same engines (for assorted, fuzzy definitions of "same"). The Falcon uses 1, and the Heavy uses 3 of them. The Soviets were famous for simply cramming more motors on to their larger rockets. If you want your speed boat to go faster, bolt on another outboard to it. There's a reason large ships have 4 screws.
Obviously there are efficiency concerns, scale concerns and overhead.
If you can get a single 4G M-Drive, it should be more efficient than two 2G drives. It should, in theory, use a bit less total space, and perhaps use less power, and cost a little less. 4G drive may use less total surface area for the thruster.
But 2 2G drives gives you M-Drive redundancy. While it may use more total surface area than the 4G, it's in 2 pieces given design flexibility. Maybe the 2G drives are from a lower TL, and more available.
Two TL 11 2G drives may have service benefits over a TL13 4G drive.
Sure, it would be nice to have that perfect, 17.2 MW fusion power plant, but Harbor Freight has an off the rack 5MW model that they sell in a blister pack next to the candy and gum at the register, and you can just bolt them to the deck next to each other.
Even the US Navy uses "stock", "off the shelf" drives (the Ticonderoga has two GE turbines, who sells them to lots of folks).
Similarly you can use "jump modules" or something to make build jump drives out of.
I think "stock drives" can offer a lot of nuance. The LBB problem is that they're monolithic.