A series elastic actuator (SEA) is a promising actuation method in force control applications that intelligently interacts with environments. The SEA is characterized by a spring placed between a load and an actuator, which is an electric motor in most cases. Since the spring plays the role of a transducer between position (i.e., the spring deflection) and force, it is able to control the output force (torque) precisely by utilizing typical position control methods. However, the force control performance of the SEA is considered to have limitations due to its elasticity, and thus, to be inferior to rigid actuators in terms of bandwidth. This paper proposes that the force control performance of the SEA can be improved by exploiting the dynamic model of the SEA. To this end, the SEA is modeled and analyzed utilizing the two-mass dynamic model, which is a well-known and widely accepted model of the flexible system. The disturbance observer and feedforward controller are introduced as the model-based control algorithms for the SEA to achieve the high-precision force control. In addition to high-bandwidth force control, the proposed controller can address the robust stability and performance against model parameter variance and exogenous disturbances. For the analytic and quantitative assessment of the proposed force control system, the dynamic characteristics of an SEA under various control algorithms are analyzed, and the experimental results are provided for an actual SEA system in this paper.