https://physics.stackexchange.com/questions/107091/soft-question-the-best-way-to-learn-physics-road-map was marked as a duplicate to How to learn physics effectively and efficiently when I have clearly mentioned why it is not, in the question itself. Will Qmechanic or someone else please explain why it was closed? Do the moderators even care to read the question before closing it?
QMechanic's comment really says it all. You are correct that your question is not an exact duplicate (and the moderator made this explicit, so accusing them of not reading the question is hardly the type of constructive discussion we expect in meta).
However, your question is a very poor fit for this site, for a number of reasons. This format works best when questions have something like a unique answer, when there is some sort of objective criterion to determine whether a given answer is correct, and hopefully one to determine which of two answers is better, and when expected answers contain facts and experiences instead of opinions. Your question fails miserably on all counts.
First of all, no one, at all, ever learns physics 'in order'; everyone learns a few things at a time. Secondly, there isn't really any way to tell one way is better than another. Voting becomes meaningless in such a question. Thirdly, one never learns all of physics, and you will be hard-pressed to find someone who can tell you both in what order to take quantum optics and quantum information, and what plasma courses to take befor learning astrophysics. Relatedly, you yourself won't learn all of physics, so that large parts of any answer would likely be irrelevant. Moreover, since you presumably have already started learning, what would you do with the answers? What if I told you that the best way is to learn math all the way up to functional analysis and then derive all of each field from its basic principles? Would you implement that?
For more general guidance, you should refer to the Meta question Good list, bad list, and to the blog post Good subjective, bad subjective. In this site, list questions are unlikely to prosper unless you make very specific requirements of the answers, and take care to avoid all the pitfalls I mentioned and in those two links.
Finally, let me clearly thank QMechanic for closing that question in a timely fashion before more people became invested in it. It was an open-and-shut primarily-opinion-based, too-broad question.