Typical all-steel buckling-restrained braces (BRBs) usually exhibit obvious local buckling, which is attributed to the lack of longitudinal restraint to the rectangle core plate. To address this issue, all-steel BRBs are proposed, in which two T-shaped steel plates are adopted as the minor restraint elements to restrain the core plate instead of infilled concrete or mortar. In order to investigate the factors that characterize the hysterical responses of this device, different finite element (FE) models are developed for the specific context. The FE models are developed based on nonlinear finite element software, which incorporate continuum (shell or brick) elements, large displacement, and deformation formulations. In these FE models, two different steel constitutive models are adopted to precisely reproduce the cyclic response of the BRB component. Meanwhile, comparisons between the numerical and experimental results are conducted to validate the effectiveness and accuracy of the robust FE model. The agreements between experimental observations and numerical predictions demonstrate that the FE method could be utilized for in depth parametric analysis. Furthermore, BRBs with detailed configurations can provide excellent hysteretic behavior and seismic performance through the optimal design process.