Physically, the coercivity of permanent magnets should scale with the anisotropy field of ferromagnetic compounds, H_A; however, the typical coercivity values of commercial polycrystalline sintered magnets are only ~0.2 H_A, which is known as Brown’s paradox. Recent advances in multi-scale microstructure characterizations using focused ion beam scanning electron microscope (FIB/SEM), aberration corrected scanning transmission electron microscopy (C_s-corrected STEM), and atom probe tomography (APT) revealed detailed microstructural features of commercial and experimental Nd-Fe-B magnets. These investigations suggest the magnetism of a thin layer formed along grain boundaries (intergranular phase) is a critical factor that influences the coercivity of polycrystalline magnets. To determine the magnetism of the thin intergranular phase, soft X-ray magnetic circular dichroism and electron holography play critical roles. Large-scale micromagnetic simulations using the models that are close to real microstructure incorporating the recent microstructure characterization results gave insights on how the coercivity and its thermal stability is influenced by the microstructures. Based on these new findings, coercivity of Nd-Fe-B magnets is being improved to its limit. This review replies to the most frequently asked questions about the coercivity of Nd-Fe-B permanent magnets based on our recent studies. Graphical abstract