In brief: The recipe is not the ingredients, nor is it the act of preparation. The recipe sits there on the page, and it doesn't taste good, either. A great deal of formal thinking has gone into the relationships between our knowledge of how to perform "acts" upon "things", the knowledge we possess about those acts and things, and what those acts and those things actually are. It turns out that we can assign certain attributes to "acts" and "things", but this only tells us how our performances are likely to unfold, and tells us next to nothing about the acts and things themselves.
Theories like Quantum Beyesianism and the problems it addresses raise a certain point that seems to have gotten lost. I have taken quite a long time to come to it myself simply because our society seems to be purposefully hiding it, or hiding from it. Hume and Kant had a discussion on this subject, and everyone seems to think that Kant won. He didn't. He lost miserably. Kant had to pull the old "my Dad will beat you up" argument out to really answer Hume's objections. Hume had said that there is no rational argument that can connect our perceptions to each other, nor to any substance, nor to any substantive cause. Kant responded with his argument of "a priori" and with a faith in God that would not have created a universe without some very deep meaning at its very core.
Kant's "a priori" argument works very nicely at the scale of normal everyday life, and the appearances of objects with which we feel familiar. I would argue that these "a priori" apply to our instinctive and embodied performances, but they cannot be applied to the models of the Universe that we carry in our heads. We have evolved in such a way as to have certain sense perceptions and abilities that cause us to survive, to persist, to pass on our genes and our memes to future generations, but this does not necessarily mean we currently or ever will have the tools to understand our Universe in full; not through the approach that we are using right now, anyway.
Kant's appeal to the merciful and intelligent nature of a Supreme Being worked nicely back when he used it. Who would object and seem like a godless pagan or atheist? That argument simply doesn't carry the weight today that it did then, however. Kant asks us to take the eventual full understanding of our Universe on faith alone, which is a terrific theological argument, but is rather lacking as a philosophical one.
Without our intervention in a pair of events, A and B, we can only observe the statistical correlation of A occurring prior to B, and there need not be any causal relationship. A and B can be "caused" by some other entity, the way a cough and a fever are caused by an infection. It seems unbelievable, but we can't ever know for certain that there isn't some such factor or factors at work, and so we cannot ever say that A causes B. The only model of causation for which we have such knowledge is when we cause an event to occur by willing it, and performing a set of tasks to make it so.
Further, our own bodies and minds, through which we perform such tasks are both composed of complex and fundamental systems beyond our present understanding, such that we cannot truly understand our performances, either. I believe that we are compelled to create a form of science that transcends the subject/object duality that we seem mired in today. We already face the prospect of a science which must largely make use of statistical correlation in preference to causation, and in the future we may need a multiplicty of scientific methodologies for the purpose of exploring a variety of phenomenological inquiries. When would you like to begin? Now, or when an inevitable crisis forces the issue?