Virtual modeling can connect abstract disciplinary knowledge with experimentation, problem solving, and reflective activity, yet multi-institutional evidence on its joint contribution to creative thinking and digital-pedagogical competence in pre-service teacher education remains limited. This multi-institutional quasi-experimental study evaluated a structured virtual modeling-based instructional system implemented from 2023 to 2026 at three higher education institutions in Uzbekistan. The intervention integrated 2D/3D simulations, model-based learning, problem-based and project-based learning, SCAMPER, Peer Instruction, Mind Mapping, reflective tasks, and a creative problem-modeling sequence. Aggregate baseline distributions were comparable between experimental (n = 226) and control (n = 222) groups, χ²(2) = 0.41, p = 0.814. At final evaluation, experimental (n = 231) and control (n = 225) distributions differed substantially, χ²(2) = 57.72, p < 0.001, Cramér’s V = 0.36. The odds of attaining at least a medium developmental level were higher in the experimental group (OR = 4.93, 95% CI 3.22–7.56), with a consistent association across institutions. These findings support virtual modeling as a pedagogically integrated approach for strengthening creative-professional and digital-pedagogical development in pre-service teachers, while causal interpretation remains limited by the non-randomized design.