I’m not sure I understand your question, and possibly the problem is in the set of assumptions either you or I have.
In the first place, nuclear decay isn’t “due to photons”. Many forms of nuclear decay don’t involve photons at all. Alpha emission results when an alpha particle (two protons and two neutrons, a helium-4 nucleus) is emitted. There are several forms of beta decay, but your classical form involves emitting an electron and an electron antineutrino. Nuclear fission results in the emission of several particles, and of gamma rays, which are high-energy photons. so nuclear decay can involve and require photons, but it isn’t “due to them” as a cause.
To say that photons don’t exist, “only waves” gets to the heart of wave-particle duality. Photons (and electrons, and other particles) have properties of both particles and waves. That’s kind of hard to get away from. Just when you think you’ve pinned down the identity to one type – light reflects, refracts, and diffracts the same way waves do, but it’s hard to see how particles would do that – you get other features that suggest the other type, such as the quantization of energy that suggests particle nature.
I suspect part of the confusion is the concept of a particle existing only at a specific point in space, which I think you hold, based on your phrasing (“where one photon propagates”) But electrons and protons have sizes (or at least characteristic lengths), despite their wave nature. How big is a photon? I can make a good argument that a photon isn’t a point particle, nor even a hard sphere, but more like a fuzzy ball that trails off gradually, much like the true shape of atomic orbitals.
Consider the single-slit experiment. People get hung up on a false dilemma with the double-slit experiment - “Which slit does the photon go through?” But consider the single slit-- there’s a characteristic diffraction pattern for a single slit. The narrower the slit, the broader the pattern. You can do an experiment where you can be certain that the flux rate is so low that only single photons are pasing through the slit, yet the same characteristic diffraction pattern results. It’s always dangerous to try to visualize quantum events with human-sized models, but I think it’s fair to say that the photon effectively “senses” the full width of the diffraction slit, which is evidence that the full extent of the photon must extend out to pretty macroscopic distances, since the individual photons contributing to the pattern aren’t communicating with each other – the information must be there, sensed by each photon. It’s like a dust bunny from under your bed with tendrils that extend to a great many times what you think its size ought to be.
Physicist view wave packets not as single-frequency waves nor as tightly-defined sharp-edged balls. Your photon (and electron, and other particles) shares not only wave and particle characteristics, but also acts like a wave packet made up of many components of differing wavelength.
Finally, "My theory is that it can be seen anywhere until it is “consumed” by the observer (or a wall or whatever). " doesn’t make a lot of sense. The photon can’t be seen anywhere until it is “consumed”. And onced you consume it with your detector, that’s it.By my mental picture, the trailing off “wings” of the photon extend very, very far from where the photon probably is, and in that sense it is “anywhere”, but all locations are not equally probable. In that sense, the beam of light can be said to propagate through a certain, well-defined region, even if there is enough of a beam outside that area to allow for diffraction effects. A lowest-order laser beam is demonstrably a well-defined and confined beam, with 99% of its energy lying within a radius of three sigmas of the center, even if 1% might lie outside that region.