Three-dimensional printing (3DP) technology has been increasingly applied in health profession education. Yet, 3DP anatomical models compared with the plastinated specimens as learning scaffolds are unclear. A randomized-controlled crossover study was used to evaluate the objective outcomes of 3DP models compared with the plastinated specimens through an introductory lecture and team study for learning relatively simple (cardiac) and complex (neck) anatomies. Given the novel multimaterial and multicolored 3DP models are replicas of the plastinated specimens, it is hypothesized that 3DP models have the same educational benefits to plastinated specimens. This study was conducted in two phases in which participants were randomly assigned to 3DP (n = 31) and plastinated cardiac groups (n = 32) in the first phase, whereas same groups (3DP, n = 15; plastinated, n = 18) used switched materials in the second phase for learning neck anatomy. The pretest, educational activities and posttest were conducted for each phase. Miller's framework was used to assess the cognitive outcomes. There was a significant improvement in students' baseline knowledge by 29.7% and 31.3% for Phase 1; 31.7% and 31.3% for Phase 2 plastinated and 3DP models. Posttest scores for cardiac (plastinated, 3DP mean ± SD: 57.0 ± 13.3 and 60.8 ± 13.6, P = 0.27) and neck (70.3 ± 15.6 and 68.3 ± 9.9, P = 0.68) phases showed no significant difference. In addition, no difference observed when cognitive domains compared for both cases. These results reflect that introductory lecture plus either the plastinated or 3DP modes were effective for learning cardiac and neck anatomy.