The free energy principle has far-reaching implications for cognitive science. In fact, the free energy principle seeks to explain everything related to the mind. Due to this explanatory ambition, it has been deemed preposterous by researchers. Jakob Hohwy challenges the opponents of the free energy principle and its applications by demonstrating that this framework is everything but preposterous. Rather, he compares the free energy principle with the theory of evolution in biology. The theory of evolution is not discarded due to its unifying power; and the free energy principle shouldn’t be either. In this paper I will present a negative as well as two positive theses: first, the free energy principle will be contrasted to Bayesian theory with regard to the degree of unification they offer. I will argue that the unification resulting from the free energy principle can be regarded as stronger since it attempts to empirically ground its conclusions in the brain via neuroscience and psychology. The negative thesis consists in the suggestion that one major flaw of the free energy principle, taken as such, lies within its explanatory power. As a result of being a functional theory, the concepts it employs are also functional. Yet functional concepts, at least when it comes to explaining the brain and cognitive phenomena, do not explain how a certain phenomenon actually works, but rather how it should work. To improve this situation, the second positive thesis of this paper makes use of a suggestion by Piccinini & Craver (2011), namely that functional analyses are mechanism sketches, i.e., incomplete descriptions of mechanisms. In other words, functional concepts (such as precision) must be enriched with mechanistic concepts that include known structural properties (such as “dopamine”) in order to count as a full explanation of a given phenomenon. The upshot of this criticism lies within the free energy principle’s potential to act as a heuristic guide for finding multilevel mechanistic explanations. Furthermore, this paper will not advocate that functional concepts should be fully replaced or eliminated, but that functional and mechanistic descriptions complement each other.