Computer-assisted techniques and the strategy of combining chemical substructures were used to derive small modified peptides as potent angiotensin-converting enzyme (ACE) inhibitors. A multivariate image analysis applied to QSAR (MIA-QSAR) model was constructed using 87 compounds with experimental pIC50 values, demonstrating good predictive performance (calibration r²=0.909, leave-one-out q²=0.604, external test rtest²=0.779). Novel compounds were designed by combining substructures of highly active derivatives, with those containing SH or SeH groups showing greater promise (e.g., compound F with a predicted pIC50 of 8.40). Docking studies confirmed the interaction of these compounds with the ACE active site, and docking scores correlated well with predicted pIC50 values (R²=0.823). Pharmacokinetic evaluations indicated that the proposed compounds possess satisfactory drug-likeness and improved properties compared to captopril. In summary, small modified peptides with favorable potency and pharmacokinetic profiles were identified, providing a basis for the synthesis of novel selenium-based ACE inhibitors.