Yes, it is completely true, and not in a probabilistic way, either. A model is a mathematical construction which relates observations to predictions. However, t

Yes, it is completely true, and not in a probabilistic way, either. A model is a mathematical construction which relates observations to predictions. However, there is no epistemic basis by which predictions can be turned into observations; we can never ultimately draw conclusions just based on inferences. Perhaps you have heard of the idea that "the map is not the territory" [0]. Models can never be exactly descriptions of reality, not without some sort of special rationale and argument from the outside of reality. In particular, QM models aren't 100% right. Gravity is missing almost entirely from the model (!) and there are some glaring experimental discrepancies, particularly around the vacuum catastrophe [1]. We know that the combination of QM and relativity gives a hybridized model that cannot work at all scales but explains things like the color of gold [2], so we know that there ought to be a single unified model which does work at all scales and has the same explanatory power. [0] https://en.wikipedia.org/wiki/Map%E2%80%93territory_relation https://en.wikipedia.org/wiki/Map%E2%80%93territory_relation [1] https://en.wikipedia.org/wiki/Cosmological_constant_problem https://en.wikipedia.org/wiki/Cosmological_constant_problem [2] https://en.wikipedia.org/wiki/Relativistic_quantum_chemistry https://en.wikipedia.org/wiki/Relativistic_quantum_chemistry