Obviously each conceptual frame or paradigm by its nature limits what can be investigated in the first place. And has its own methodology, assumptions, axioms, etc. which may be similar to but different from other enacted paradigms. Hence we get the likes of Wilber's principle of nonexclusion. Like you he maintains though that nonexclusion can be superceded by enfoldment:
“However, if one practice or paradigm includes the essentials of another and then adds further practices—such that it 'enfolds' or in cludes the other—then that paradigm can legitimately be claimed to be more integral, which is the enfoldment principle” (Excerpt D, 2).
Hence our discussion of metatheory and metaphilosophy, which crosses paradigms to see where they intersect and where they are incommensurate. And which often exposes the hidden premises and/or axioms upon which a given paradigm is based and thus prejudicing its view.
I also accept supervenience, that we cannot reduce one paradigm to another. But obviously Kuhn himself recognized that we don't just wait till the people that promote on paradigm die off to be replaced by another, but that a better (for now, not forever) paradigm emerges from the previous one and replaces it. Hence progress, not just lateral difference. Or as you say, an enfolding meta-paradigm that better contextualizes the others.
In that sense I think Knox, who crossed disciplines and/or paradigms, offers not an incommensurate view of archetypes but a better enfolding cross-paradigm. It seems to me is that what you are doing is reducing both image schema and Knox's recontextualization to one paradigm, like to the physical for image schema, and to analytic psychology for her archetypes. Granted I found some valid cross-paradigmatic criticism of L&J's version later in the real/false reason thread, particularly with regard to the downward causational changes on image schema via culture. And with the notion of the extended mind thesis. In this thread I've even used the latter to extend the notion of image schema into the individual-environment assemblage instead of limiting it to the inside of the individual.
So we need more than just what an archetype is for the four (worlds) paradigms; we need a meta-paradigm that makes sense of the connections or relations between those four paradigms and can make better (for now, not forever) cross-generalizations and pattern recognitions on all four. And one of those better replacement recognitions is going postmetaphysical. Which is one way Knox supervenes on Jung, the latter which unquestioningly had a metaphycial bent to his archetypes. And which metaphysical bent must not be transcended and included but transcended and replaced. (See the basic and transitional structures thread.)
But returning to the issue of reducing everything to image schema, that's not what I suggested. Nor does L&J for that matter, since obviously they expound at great length on how it sets the ground limits and parameters but that stuff nonetheless emerges from them at higher levels, like abstract thought and metaphor. The focus on image schema is merely to embody later emergences so as to avoid the pitfall of a metaphysical worldview and philosophy that has the hidden axiom of Platonic ideals and/or Aristotelian set categories rooting their paradigms. That is the better (for now, not forever) replacement axiom, or at least one of them, of a postmetaphysics. And highlights the difference between real and false reason.
And not only that, but how the image schema themselves are transformed into causual and nondual states and stages of human development via the use of the more emergent and developed functions like the abstract ego, another of my points and explored at length in some of the referenced threads. That particular cross-paradigm exploration argues against the notions of a false reasoning that just keeps getting more and more complex as in the model of hierarchical complexity, itself based on the above metaphysical axioms. Whereas a real reasoning, even the bastard reasoning mentioned recently in this thread, more easily accomodates the kind of general complexity and general economy noted in the pomo and complexity thread. All of which are advancing through a Kuhnian progression and/or revolution of paradigms, scientific and otherwise.
Expanding on my last point, while the 4 worlds paradigms are to some degree incommensurate or irreducible, they are nonetheless not static and unchanging like Platonic ideals or Aristotelian categories. They are generative (en)closures, so to speak, that are in assemblages, being both closed within their own autonomous substance and open to other suobjects, and to the changes that occur therefrom. So as but one example even the physical, in response to the mental, in response to the cultural, all interact and change the very endo-structure of each. Endo-structure, even though defining a separate and irreducible autonomy, is nonetheless contingent and changes throughout evolution, both individually and culturally.
Even image schema as the basic endo-structural elements of thought change in response to downward causation from mind, environment and culture. Even DNA changes this way! Still, image schema, though changed, are still an elemental endo-structural format, thus setting general parameters. Those parameters are then further expanded or constricted depending on the hyperobject assemblage, which itself depends on its suobjective parts and its own elemental endo-structure. And as I've speculated in this thread, as root of the 'universe' hyperobject is itself differance, itself arrived at via a bette (for now, not forever) metaphilosphical or postmetaphysical transcendetal deduction.