CB₁ and CB₂ receptors belong to the large family of G-protein coupled receptors (GPCRs), being involved in a wide variety of signal transduction processes. In this context, CB₂ selective compounds were described in the literature to be active in different neuropathic and inflammatory pain models, also showing beneficial effects as neuroprotective agents. Indeed, CB₂ proved to be up-regulated in reactive microglial cells in Alzheimer's disease (AD) and Huntington's, suggesting a promising therapeutic panorama for CB₂ inverse agonists/antagonists. At present, the development of new selective CB₂ antagonists is prevented by an unclear depiction of the reference antagonist SR144528 binding mode. Indeed, a few number of models concerning the CB₂ receptor antagonist binding site have been proposed, leading sometimes to contradictory results. In this context, a specific hCB₂ ligand-based homology model in the presence of the antagonist SR144528 was built. Notably, the refined model also allowed us to explore the pyrazole scaffold as prototype for CB₂ ligand recognition and also to elucidate the CB₁/CB₂ structure–activity relationship of an in-house series of analogues we previously published. The derived information is expected to be a useful tool for guiding a much more focused and reliable CB₂ antagonist design.