It may well be that multiple explanations are consistent with the most basic phenomena. The criterion for the reality of the theory is whether it does better than random (its predictive power, equivalently how well it can compress the observational data). We can try to find the simplest explanation, but we can never know if we have it [2]. We can stop thinking of theories as being real or not in a binary sense, and merely ask how well they compress the data. Of course, different theories can achieve the same compression by compressing different aspects of the data! Your pattern can look like my noise :-)
Luckily and interestingly, the physics of the macroscopic world is in many cases effectively independent of the underlying microscopic rules [3], with the details of the microscopic physics being screened off as complexity emerges.
Personally, I like to think of abstraction as a kind of hierarchy of hardware + software: at each level, some useful ontology is picked out for stable patterns to emerge (e.g. the software of "fluids" emerging from the hardware of quantum "particles"). Layers of software at one level become the hardware substrate for the next rung up the abstraction ladder.
Luckily, we get to choose the software that runs on us, so don't forget to vote this fall ;)
[1]: https://ruccs.rutgers.edu/images/personal-zenon-pylyshyn/cla...