This is just the copenhagen interpretation which says nothing about the physical ontology of the light before it ends up as a point on the screen. Unless you are saying the probability amplitudes are physical things-in-themselves rather than calculational models. (side note - Feynman did not think of a probability amplitude as a physical thing-in-itself)

But did Feynman think anything as a physical thing-in-itself? Feynman repeatedly argued that that physics is whatever the instruments measure, generalized by whatever computational models match the measurements, and he was happy to analogize to metaphorize whenever convenient to describe an aspect of a phenomenon ("photons jiggling")

One good thing about Feynman was he was always clear on what was experimentally observable and is what it is and what is theory where you can make up anything that fits the experiment.