The number which Shannon's estimate arrived at, which exceeds the estimated number of atoms in the observable universe, was for the number of different games, not positions. The latter estimate is very much smaller than the number atoms in the observable universe.
On the subject of the uncertainty principle, we can know both the position and speed of an elementary particle accurately. It is position and momentum, and by extension, velocity, which form a complementary pair of variables.
Claude Shannon once calculated that the number of chess positions is greater than the number of atoms in the universe.
Is the comparison between chess and nuclear physics in place? Sure it is. People get PhDs in nuclear physics in 8 years, they build nuclear reactors etc, however, it takes even longer to learn to play chess at GM level. Nuclear physicists have to remember much less patterns than chess players. Chess in not natural phenomena, but it may be studied as such, using calculations, probabilities and so on. It should be studied at Universities, alongside math and physics.
Just like nuclear physics, in chess we deal with unknown principles, invisible entities - we have to use imagination and logic in order to calculate. Remember the indeterminacy principle in quantum physics? Just like we know the current position of a particle but not its speed and future positions, so we know the current position of a chess piece but not its future positions. These are just probabilities. It's very difficult to calculate chess positions, did I mention that?
Therefore, make sure to buy courses on every important position, opening, tactics and so on, play many online blitz games and puzzles (constructed or taken from unknown games), and of course - easily beat 2200 elo bots on a weak CPU. That's how you will gain proper understanding of the game.