Breast cancer continues to be one of the most prevalent and deadly cancers in female populations of the world, underscoring the need for early and accurate diagnosis. Mammography is the primary diagnostics tool for screening in breast cancer diagnosis, and through recent advances in deep learning, automated diagnostic support systems have been developed for the use of mammographic images. This work introduces a deep learning-based approach for classifying breast cancer using digital mammography images. Features from mammographic images were extracted using two of current state-of-the-art, pre-trained Convolutional Neural Networks (CNNs): EfficientNetV2B0 and ConvNeXtBase, using over current best features derived from pre-processed images. The models were assessed independently and then jointly. ConvNeXtBase achieved an accuracy of 98.60% while EfficientNetV2B0 had an accuracy of 98.67%. The joint model had a classification accuracy of 98.33%. The study demonstrates the potential of the proposed methodology and the possibility that mammography and automated diagnosis tools could improve automated diagnosis of breast cancer.