Very enjoyeable essay! I liked how you explained difficult concepts like the classification of the wavefunctions by spin and mass in terms of representations of the Poincaré symmetry, made by Wigner and Bargmann, the role (or absence of such a role) of the observer in various interpretations, decoherence, the measurement problem etc. I like this statement "the objects themselves can never be known. Only the relations these objects have with each other shows as properties of the objects" :). I also agree with "the applicability of the basic concepts, that define a physical theory and describe the laws, might depend on the environment and hence the laws themselves might depend on contingencies". This can be interpreted in two ways: that the laws really change, or that our formulations change. For example, does the spontaneous symmetry breaking with the Higgs mechanism represent a change of constants considered before as ... constant, like the particle masses and coupling constants, or this just represent that they depend on the state, and the state transition is not changing the fundamental laws, only the laws we used to think as fundamental? (edit: I see that toward the end of your essay you wrote "The richer theory might emerge from the poorer one by symmetry breaking.")
I liked the mention of the WAY theorem. I'd just want to make the observation that, if the Hamiltonian doesn't change and the symmetry breaking happens just at the level of the system (as in Higgs mechanism), the constraints imposed by the WAY theorem don't change. Because the theorem takes conservation laws as an indication of unitary evolution, which has to be true during the pre-measurement phase, when the system is supposed, e.g. in von Neumann's formulation, to branch into a superposition corresponding to different outcomes. So the symmetry of the system has no relevance, because the theorem uses mean values of the conserved quantities to check unitarity, and not eigenvalues. For these constraints to change in time, the Hamiltonian itself has to change, and I understand that this is a thing that you want to propose.
Another thing I liked was the mention of Poincaré's "Science et l'Hypothèse", which, together with "Science et méthode", were among my favorites when I started to be interested in physics as a kid. However, I never agreed with him that there's actually no difference in what geometry you choose. Take his representation of hyperbolic geometry in the Poincaré disk. If you take the disk with the Euclidean metric, you'll have to admit that the objects deform as they move. If you take the hyperbolic metric, they are rigid w.r.t. the hyperbolic geometry. So, in such a world, you'll have these two theories, one in which the objects are not rigid, but they deform as they move in a conspirational way, that makes them behave "as if" they are rigid w.r.t. a hyperbolic metric which is just a construction. The other one in which these "coincidences" of zero probability are explained by the symmetry of the hyperbolic geometry, which makes them have probability 1. Klein's insights in his Erlangen program are really genius, and the entire physics, from relativity to quantum mechanics and Wigner's classification and gauge theory, is indebted to these ideas. We can keep an open mind that maybe these "coincidences" can be explained in a different way than because of the symmetries of the geometry, but this is not the same as saying that they are undistinguishable. I think this idea made Poincaré miss the opportunity to become the main author of the theory of relativity and Minkowski spacetime. But, even if it is more in the spirit of Occam's razor to pick the theory which requires less conspirational fine tuning to explain the "coincidences", Poincaré's observation is important, in the sense that it is always possible that a better explanation, with completely different rules, will explain the same data with fewer hypotheses and adjustments to explain the apparent symmetries.
I like the way you take various ideas and make them converged to your proposal of semantically closed theory, which makes sense. I wish you success in the contest!