There is nothing wrong with correcting previous theories. Most of the theories were properly constructed, but critical parts could not be tested for with the equipment available at the time. Many of the phlogiston equations, for example, could still be used today if we just subtract the phlogiston from one side of the equation and add a corresponding amount of oxygen to the other side. Becher had to make a guess about what changed when substances burned. His guess was wrong, mainly because he didnt capture the gasses released during the burning, so he thought the fuel lost weight, and therfore lost an element. When Priestly first isolated oxygen, he called it "dephlogistonated air," until he realized that it was pure "anti-phlogiston."
The thing is Benjamin Franklin made exactly the same kind of wrong guess when he was experimenting with electricity. And we still use his backward equations, today. That is why electrons carry the negative charge, and the current is said to flow in the opposite direction to the electrons that carry it.
Theories are models, and no model can properly model every aspect of the reality. Scale models of machines and houses have square/cube ratio problems. Flat maps of the world are always distorted in one way or another, etc. But as long as you know the model's limitations, and the thing you are interested in either does not involve those limitations, or those limitations can be compensated for, even an obsolete model is as competently useful as the latest model.