You don't even need to state whether you assume physicalism as your world view for this question.
Even a pure physicalist would be very used to modeling different parts of the world in different ways, with different axioms.
As an example: if a physicalist would model the behaviour of large groups of people - for example, a statistican working in an insurance company - they would work with certain models, axioms, rules and so on and forth which are tailored specifically to the domain. Even if they believe that humans can just be reduced to their atoms (nay, quarks...), they would never use the axioms ruling in the quantum world to calculate the finances of their firm. Granted, they would maybe not contradict real physical axioms, but they would be very separate; mostly having nothing much in common at all.
As a physicist (not physicalist), you would be very used to working with different models and outright contradicting axioms if you are looking at different scopes. For many areas of our daily life, the Newtonian classic mechanics are just fine, and enough to send rockets to the moon; even though we know that they are ultimately wrong. As a practical chemist, you can work and be successful all your life modeling atoms or molecules the old-fashioned Rutherford-Bohr way (which is still helpful for didactic purposes, but physically pretty much wrong), and completely ignore most if not all of quantum theory.
Modelling fluids and gasses is so utterly complicated that we don't even try - our best efforts are very very crude approximations. And still we build great vehicles that fly at many machs.
So, as even intra-physicalistically speaking it's perfectly valid or even necessary to have different axioms for the same areas of the world, very much the same must be true for your purported distinction between physical and mental; the very nature of the distinction enforces --some-- kind of difference, and that difference must lie in axioms (and rules and assumptions and so on and forth).