The cmu doesn't "have" to be the strongest part, but there's decades of conventional practice to prove that using high-strength grout in the wrong application can be detrimental to the entire assembly. One of the biggest issues is that it lends to additional shrinkage cracking through the units, sacrificing the strength of the non-re-enforced block, which sacrifices the assembly as a whole. Also, CMU are close to the perfect environment for curing, so it's not uncommon to see a 2500 psi rated mix exceed 3500-4000 psi in actuality in the wall due to the amount of moisture surrounding the pilaster.
Assuming that 4-4.5k psi pea gravel concrete is an acceptable replacement for grout in cmu walls is irresponsible, as there's plenty of info available out there to debunk the idea. Most importantly, pea gravel concrete will segregate long before grout will at the high slump that is required for grouted assemblies.
As for other other examples, they're not applicable to the OP's situation.
Lastly, YES, I have seen (and been bound to adhere to) minimum & maximum grout compressive strengths on the same job. I would assume the reasoning for those specs is that the engineer had seen his share of high strength grout failures with his own eyes........