Lol, good luck "accurately" mapping our irregular oblate spheroid on to a globe (i.e. a regular sphere). You're correct that map projections inherently make concessions and introduce inaccuracies. But a globe would just require a different kind of projection in this context. And that projection hides more subtle inaccuracies. Globes are not an answer to the problem of map projection distortion.