The reductionist’s approach to human performance modeling provides a framework for developing modular human performance models that can be aggregated to represent more complex human behavior. A simulation experiment was conducted to explore the use of this approach to mimic the secondary-task technique in a computerized model for task workload assessment. Two task network models were independently created to represent an Air Traffic Control (ATC) and a visual Bakan task respectively. The two models were then connected in parallel to represent operator behavior and assess dual task performance. The Bakan task was assigned as a secondary task and performed concurrently with the primary ATC task. The model generated performance predictions for two tasks under both single and dual task conditions. The results produced patterns of performance that could be explained by the secondary-task technique. The study demonstrated the feasibility of using the secondary-task technique for assessing task workload in human performance models.