Jiles-Atherton　(JA)　model　has　been　widely　used　to　describe　the　magnetic　hysteresis　property　of　Terfenol-D　material.　However,　the　accuracy　of　the　classic　JA　model　is　not　convincing　any　more　when　it　is　employed　to　predict　the　magnetizations　under　various　amplitudes　of　the　applied　magnetic　fields.　This　paper　measures　the　magnetic　field　and　magnetization　indirectly　and　approximates　the　measured　results　to　the　best　of　the　classic　static　JA　model.　Then　it　shows　that　the　JA　model　fails　to　describe　the　magnetic　hysteresis　effectively,　especially　in　computing　the　hysteresis　width.　The　high　computational　error　is　caused　from　inappropriate　value　of　the　pinning　parameter　to　quantify　the　average　energy　required　to　break　pinning　site.　Then　the　parameter　is　modified　from　the　constant　value　to　an　independent　variable　on　the　field　amplitude.　From　the　comparisons　on　the　results　of　the　maximum　magnetization,　maximum　magnetization,　residual　magnetization　and　coercivity,　the　performance　in　describing　the　hysteresis　property　of　Terfenol-D　is　obviously　promoted　by　the　improved　model　apparently　compared　to　the　classic　one.　©　2024　Elsevier　B.V.