Scientific graphs and images play an important role in data analysis and the understanding of concepts. However, they are primarily explored visually, which limits their accessibility for people with visual impairments. In an effort to create tools to reduce this limitation, a Python-based software was developed to generate 3D models from 2D images, which also include associated text in Braille. These models enable tactile exploration of scientific data representations on a Cartesian plane, similar to how they are typically presented for visual inspection. The printed tactile models were evaluated by people with and without visual impairments. The former considered the tactile models a valuable support tool for interpreting representations and accessing data beyond touch alone, as the experience was enhanced by an added sonification system, turning it into a multisensory experience. For people without visual impairments, the experience also provided learning opportunities. This article presents the first qualitative results on the use of multiple senses for data analysis and pattern recognition across various types of signals.