Transition state theory tells us that the rate constant of an elementary step is $$k_\mathrm{r} = \frac{\kappa k_\mathrm{B}T}{h}\exp\left\{\frac{- \Delta G^\ddagger}{RT}\right\},$$ where $\Delta G^\ddagger$ is the activation free energy of the step, i.e. the difference between the Gibbs free energy of the transition state and the reactant state.
Then the rate law would be $$r = k_\mathrm{r}[A]^a[B]^b[C]^c\dots,$$ where $A$, $B$, $C$, ... are all the reactants that come together in that elementary step. So far so good.
But I have come across reactions where changing the concentration of a reactant (by a large amount) in a multi-step reaction seems to change the rate determining step of the whole reaction.
The explanation offered was that changing concentration changed the Gibbs free energy of the reactant or intermediate states (as Gibbs free energy depends on the concentration in solution). The states which contain the species (whose concentration is changed) are affected whereas others are not, so the Gibbs free energies of activation are changed, which means the rate constants are changed, which can change the rate determining step.
But if concentrations can affect $k_\mathrm{r}$ then it seems like the rate law cannot be written like that.
Does this sound right, or am I overthinking this?