Computational fluid dynamic (CFD) simulations have attracted a lot of attention in the last few decades thanks to their ability to simulate complex flows in intricate geometries, such as the human respiratory system. Well-suited procedures, such as mesh validation and numerical sensibility analysis, standardized, e.g., in the aerospace community, are not yet systematically adopted in the vast literature of CFD simulations for the human airways. Here, we aim at highlighting how troublesome the exploitation of such computational procedures can be for a coupled CFD-DEM (Discrete Element Modelling) approach inside the upper intra-thoracic airways, in the framework of the SimInhale model, [1], [2], [3]. The highlighted topics, based on our papers [4], [5], are: 1. Mesh validation. A standard procedure where one refines the number of employed elements in order to obtain a mesh-independent fluid dynamical fields. The effect of different meshes on the particles dynamics and deposition data is critical, highlighting the need for further analysis or better DEM algorithms for simulating particles dynamics. 2. Boundary conditions (BC). We will show how employing different boundary conditions will bring different sub-lobe ventilations compared to experimental data, [1], [3]. The enormous effect of different BCs on the particles’ dynamics will be highlighted, bringing the attention on the need to developpe physiologically meaningful BC based on patients-specific data. 3. Secondary structures in lower-airways. In 1D models estimating particles deposition in the lower-airways, often a 1D Poiseuille like flow is assumed in the deeper airways. Secondary structures, shown in [4], [6], [7], propagate in the lower airways and thus prompting reflection on the need for 1D models to adjust the assumptions to better mimic the realistic airflow. 4. Turbulence and time-varying structures in lower-airways. As seen by many authors, [4], [5], [7], [8] , fluid dynamical fields have intrinsic instabilities inside the available airway geometry. Time oscillations in the mouth-throat (MT) region are classified as turbulence thanks to experimental and numerical data,[9]. There is a lack of understanding of these time fluctuations in the intra-thoracic airways. In our work, we showed how the intensity of these time fluctuations does not attenuate monotonically differently from the Reynolds number. This calls for experimental measurements capable of characterize their nature.