Bigger drivers = bigger soundstage

There is so much more to that than just the driver size...
Longer time ago I had a smaller room, smth like 8’4 by 14’, 10’4 ceiling, old apartment building. It could be filled by bookshelves with 6” drivers. Floorstanders with 10” were just too much. I sold the floorstanders to a friend, in his approx 16’ by 17’ by 9’6 from they are much better (even when poorly placed).
my current room is 21’ by 23’, 8’4 ceiling with 7’ by 10’ opening to above + two other openings effectively increasing the volume by some 50%. Used Lowther Fidelios for a time, poor bass with solid state, adequate with 300B SET. That’s 8” + folded rear horn.
Acquired ancient yet immaculate MBL 101c. Base drivers is approx. 9” diameter, probably 13” tall, radially pulsating elongated aluminium ellipsoid. Deepest bass was lacking, but the room was filled so much better than with Lowthers. MBLs need ML 23 to come alive, less power and control is not for them. Wanted cleaner bass, got Gradient 1.5 Helsinki. Got the clean bass (not very much of it though). Overall, almost ‘near field’ listening - truly amazing in the sweet spot, adequate background music in the whole room, yet no freedom of movement like with the MBLs.
Decided to build dipole subs “a la Celestion 6000”, got some BMS 18N862 drivers. Connected them straight to power amp (naked drivers on the floor). 18” drivers, so no heights, on the floor, so terrible early reflections - yet smth. very immediate, coherent, and “right”. Added BSS 966, lifted lows (dipole correction) and heights. Not perfect, but good enough to make me wonder why people insist on boxes, and am I really in a hurry to complete the project as intended. And I have only connected one pair of those BMS, with two per side there will be no shortage of sufficiently deep, yet clean and dry bass. But BMSs do not fill the room. I absolutely need to add omnidirectional speakers from 200 Hz (I.e., above room modes).
I will have two options for omnidirectional, MBLs and German Physiks DDD drivers. When I make ‘enclosures’ for DDDs, I will see if size matters (MBLs have much larger total radiating surface).

Inefficient, thanks

ya know, u raise a point where larger drivers may fill room better where placement offers more flexibilty.  That’s why I feel well designed larger drives offer so much. Yeah!

If I think about your particular case, presuming classic [monopole] bass reflex boxes with cone drivers for low and mid + dome tweeters, smaller drivers can move less air, thus they may sooner run out of their capacity to produce undistorted sound, esp. bass.
But harmonic distortion is only half of the story, bass has to be controlled, and clean or boomy bass does not depend on driver size.
Smaller drivers usually mean narrower baffles and better interference/diffraction and dispersion characteristics, and better directivity (less beaming as frequencies go up).
Directivity has been often ignored in the past, yet it is quite important.
For an example of deliberate choice of narrow directivity look at early “Gradient” models, for example “Gradient 1.3”.
Gradient 1.3’s dipole midrange is larger than it’s boxed woofer. It was done to increase the directivity and avoid early floor reflections.
A the same time its high frequency section is a short line array of four tweeters - again, to achieve broader horizontal plane dispersion and less in the vertical plane.
I did not mention Siegfried Linkwitz, presuming that you are familiar with his work; if not, have a look: http://www.linkwitzlab.com/ALMA'14/Sound_quality.htm
If you go more technical and dry, the measurements here: https://www.princeton.edu/3D3A/Directivity.html
can be quite enlightening.