Vehicle Routing Problems (VRPs) attract attention of combinatorial optimization research literature, for they are NP-hard. An important variant of a VRP formulation is the so-called Heterogeneous Fleet VRP (HFVRP), where the volume of transported items cannot exceed vehicle's type capacity and the fleet of vehicles is considered heterogeneous. In practice, one often has to be able to verify, whether a given sequence of items really fits into the vehicle space loading or not. This requirement is referred to as a loading capacity constraint. We assume here, that the layout of items in vehicles is also constrained. In particular, in this paper, we highlight the impact of 3D loading on VRP solutions as in real-life practices.