It is challenging to obtain in vivo or in situ experimental data from human muscles due to the invasive nature of such measurements. As a result, many investigations of human performance, surgery, or skeletal adaptation are necessarily based on musculoskeletal models. The utility of such models will depend on the question being asked and the extent to which the model is sufficiently accurate to address that question. In this perspective article, we take advantage of unique intraoperative access to the human gracilis muscle and make direct comparisons between commonly modeled parameters and those measured from the human gracilis. We directly compare muscle–tendon unit (MTU) length, optimal fiber length, and tendon slack length. Our results demonstrate that measured and modeled length parameters differ greatly. This is primarily due to the fact that slack muscle length and optimal muscle length differ greatly for the human gracilis and that models assume they are the same length.