Large transient currents and magnetic intensities are generally encountered by the superconducting magnetic energy storage systems thereby resulting in AC losses that occur during the charging/discharging durations thus, estimation of such heat loads must be known before the actual designing of cooling arrangements. Various analytical methods are accessible to approximate such losses however, not adequate to predict the same accurately. In this work, an extensive numerical model has been established to estimate AC losses among the stacked/circular coils used in superconducting magnetic energy storage applications under various transport currents (150 A to 225 A) and strong fields (1e5 A/m to 4e5 A/m). A homogenization method has been incorporated into the model stacks of high-temperature superconducting tapes (Ic=330 A @ 0 T self-field) instead of an actual multi-turned coil. The modeling has been done to achieve an anisotropic bulk equivalent for the stack keeping the overall electromagnetic identity of the model the same as that of the original multi-turned stacked geometry such that it eliminates the multiple boundaries among the geometrical layout consisting of metallic, insulating, superconducting, and substrate layers. In the present numerical model, E-J power law has been used which has considered the overcritical current densities. The results showed that transport currents affect AC losses significantly and as it approaches critical current, the AC losses increased abruptly. The novelty of the work is in the consideration of external fields which are present beforehand the actual current transportation through the coils using the homogenization approach. It has been concluded that perpendicular fields have a significant effect on AC losses compared to parallel fields. Considering the fact of high computational speed, this model can further be extended to solve more complex transient characteristics of HTS motors or generators. It has been concluded that the homogenized model can estimate average AC loss two orders faster than the original stacked model.