In physics you often find rules for modeling composite objects in terms of simple ones. For example a rigid body can be modeled for many purposes as a point mas

In physics you often find rules for modeling composite objects in terms of simple ones. For example a rigid body can be modeled for many purposes as a point mass, whose value is the rigid body's total mass, located in its barycenter. In quantum mechanics you can also form composite objects (in the most common formalism it will be a "wavefunction") and you still can determine a total fermion/boson-ness as a function of the elementary objects that form it. So yes, an atom, a proton, a molecule, a table all can be bosons or fermions once you decide to ignore some internal degrees of freedom. In fact even those subatomic particles you mention could be composite objects for all we know.