Fields are just another modelling concept. They have the advantage in that they are both local and global, so do offer a holistic approach.
A field can define a space by filling it, while locally specifying its material content. Local particles can be described as excitations and so given a contextual definition. Etc.
So if reality is holistic and systematic in its causality, we should expect a field ontology to be good at capturing that essential local~global organisation.
Of course, like any analogy, there are then shortcomings. Fields have no memory, no persistence. All is flux. So it is hard to represent history or gradients.
So classical wave mechanics has been a useful mental concept for modelling material reality. But note that soliton modelling and superconductor modelling from condensed matter physics are now also common mental concepts being employed in fundamental physics, along with network theory (as in loop quantum gravity).
And also, the essence of a "field" in any of these descriptions is that it preserves locality. Whereas QM creates a problem in that regard.