There's nothing wrong with that, but it is overcomplicating things. Wave-particle duality is a notion of the socalled "old quantum theory" which AFAIK none of the "founding fathers" of QT ever took as "the final word". Einstein always emphasized after his photoelectric paper of 1905 that he cannot stop thinking about the problem (!) of understanding radiation. He even thought it's the far more challenging problem than his General Theory of Relativity. IMHO these difficulties are overcome with the discovery of modern QT with Born's achievement of the probability interpretation of the quantum state as the key to the resolution of all mindboggling contradictions and inconsistencies of the old wave-particle-duality handwaving. According to moder QT neither a classical particle nor a classical field (or wave) description is a complete picture (in relativistic QT even a wave-mechanics picture a la Schrödinger's non-relativistic QT representation is impossible!). A quantum is a quantum. You cannot explain it by simpler ideas than QT itself. QT is today the fundamental theory of how matter and fundamental interactions are described. It cannot be reduced to something even more fundamental. Maybe that's possible in the future with even better and more fundamental theories, which might be discovered by some genius. In you photon example, it's clear that the photon is the one freely moving quantum that has the least particle-like properties ever. It doesn't even lets you define a position observable in the strict sense, i.e., said in an operational way, you cannot localize it. All you can do is to calculate the detection probability of the photon at the localization of you detector. That's all that QED provides you, and so far there's no necessity to seek for something else than QED!