There are a couple of problems with defining the concept of volume for quantum particles.

When a fundamental particle such as a photon actually interacts with something it does so as a point, as far as we can tell. In that case, we model it as having zero volume.

But in between interactions, it's smeared out over some region of space, and is described by a wave function that tells you the probability of finding that point particle at each point in space. In that case, you could define the volume in which you're X% likely to find the particle, and call that the particle's volume.

Photons are a bit tricker because of the way we describe them. Approximating them as a few wavelengths of a field is good in some circumstances, but the technical definition of a photon is a wave function over all space. I think that when photons are being sent from a source to a receiver, you end up being highly likely to see them at a certain point in space, which corresponds to the few-wavelength approximation.