Since in vivo measurement of these quantities requires invasive methods, musculoskeletal finite element (MSFE) models are widely used for simulations. There are, however, limitations in the current approaches. Sequentially linked MSFE models benefit from complex MS and FE models; however, MS model’s outputs are independent of the FE model calculations. On the other hand, due to the computational burden, embedded (concurrent) MSFE models are limited to simple material models and cannot estimate detailed responses of the soft tissue. Thus, first we developed a MSFE model of the knee with a subject-specific MS model utilizing an embedded 12 degrees of freedom (DoFs) knee joint with elastic cartilages in which included both secondary kinematic and soft tissue deformations in the muscle force estimation (inverse dynamics). Then, a muscle-force-driven FE model with fibril-reinforced poroviscoelastic cartilages and fibril-reinforced poroelastic menisci was used in series to calculate detailed tissue mechanical responses (forward dynamics). Second, to demonstrate that our workflow improves the simulation results, outputs were compared to results from the same FE models which were driven by conventional MS models with a 1 DoF knee, with and without electromyography (EMG) assistance. The FE model driven by both the embedded and the EMG-assisted MS models estimated similar results and consistent with experiments from literature, compared to the results estimated by the FE model driven by the MS model with 1 DoF knee without EMG assistance.