Computational chemistry has emerged as a sub-field in science over the last five decades, not at least due to the amazing increase in computational resources. Equally important, however, is the continuing developments of theoretical methods and in software technology. The refinement of theoretical models moves forward at a steady pace, but at times takes a radical turn in new directions. Computational chemistry methods are now integrated elements in many research fields and are routine tools for non-experts. The plethora of different models, however, forms a bewildering jungle of choices, often resulting in practitioners defaulting to the tried and true. The present contribution contains some personal reflections on the development of computational chemistry methods over the last four decades, with special focus on the development of basis sets and density functional methods.