Kirschner, Sweller, and Clark (2006) suggest that unguided or minimally guided instructional approaches are less effective and efficient for novices than guided instructional approaches because they ignore the structures that constitute human cognitive architecture. While we concur with the authors on this point, we do not agree to their equation of problem-based learning with minimally guided instruction. In this commentary, we argue that problem-based learning is an instructional approach that allows for flexible adaptation of guidance, and that, contrary to Kirschner et al.'s conclusions, its underlying principles are very well compatible with the manner in which our cognitive structures are organized.