Nowadays, different approaches have been pursued with the intent to develop sulfonamide-like carbonic anhydrase inhibitors that possess better selectivity profiles toward the different human isoforms of the enzyme. Here, we used conventional 3D-QSAR methods, including comparative molecular field analysis (CoMFA), comparative molecular similarity indices analysis (CoMSIA), and Topomer CoMFA, to construct three-dimensional quantitative structure-activity relationship (3D-QSAR) models for benzenesulfonamide derivatives as human carbonic anhydrase (hCA) II/IX inhibitors. The theoretical models had good reliability (R>0.75) and predictability (Q>0.55), and the contour maps could graphically present the contributions of the force fields for activity and identify the structural divergence between human carbonic anhydrase II inhibitors and human carbonic anhydrase IX inhibitors. Consequently, we explored the selectivity of inhibitor for human carbonic anhydrase II and IX through molecular docking, and the difference of activity coincides with the potential binding mode well. According to the results of the predicted values and the molecule docking, we found that the inhibitors published in the literature had stronger inhibition on the hCA IX; based on the theoretical models, we designed seven new compounds with good potential activity and reasonably good ADMET profile, which could selectively inhibit hCA IX. Molecular Dynamics Simulation showed that newly-designed compound D7 had good selectivity on hCA IX. The findings from 3D-QSAR and docking studies maybe helpful in the rational drug design of isoform-selective inhibitors.