I've read the article. It exposes a well-known point of view in philosophy of science that has been debated at least a hundred years. It doesn't begin to cover that question at all.

Notably, the idea that a new theory simply takes existing "observations" and adds to them is historically unfounded. Indeed, many theories start as thought experiments that contradict observation and later turn out to be better models, not the other way around.

You also may or may not be familiar with the concepts of theory-ladenness and the commensurability of theories. If not, I suggest doing some reading about that so you can appreciate why those questions aren't as straightforward as you seem to believe.

[1] https://en.wikipedia.org/wiki/Theory-ladenness [2] https://en.wikipedia.org/wiki/Commensurability_(philosophy_o...