I like your clearly stated purpose that you are learning to expand the breadth of your knowledge. I had the same mindset when I began my self-learning path about a year ago, and the path gets longer as your pursue it, so I would advise changing the 3-6 months to 3-6 years, if your plans is to learn. We can both agree that we are not looking to become experts, so we can relax and go with the flow—in 30 years, who knows what will happen.
CHUNKING
My self-learning approach is chunking subject into 1-2 week blocks. Rarely less than 1-week, never more than 1-month. It’s a cyclical process that I use to give my brain time to consolidate new knowledge.
UNDERSTAND THE BIG PICTURE
When learning a new subject, I always spend my first 2 chunk sessions to understand the big picture, and in later sessions I learn the details through deliberate practice. What I mean by big picture is when you commit time to read a book, or watch a video (@1.5x), you don’t need to read in a linear order, or watch every single minute of the video. Don’t read/watch any content with “learn ABC in less than X time” in the title. The goal is to learn best practices from experts. Only challenge the status quo once you have gained the discipline.
DATA SCIENCE (2weeks)
So in your case, spend 2 weeks learning Data Science: pickup a Wes Mickiney book on Python Data Science, or find a GitHub repository with great contributors sharing their work to help you. If you get stuck on transforming your DataFrame into Matplotlib or Seaborne, stop. Go work on your art project, or in this case, let’s read.
LITERATURE (1week)
For reading - read Strunk and White or William Zinsser if you want to improve your writing. Read Walden, Gatsby, or 1984, if you want to see thoughtfulness in writing. I rarely finish an entire book because I’m more interested in the themes and proses than every details because I have limited memory and I want to ready many books. (1 week)
FOREIGN LANGUAGE (1week)
Now spend the last week writing to your foreign pen pal. Let’s say if it’s in Japanese, learn the hiragana, which is quite easy since all the sounds are romanized, and afterwards you can use the Japanese dictionary, instead of Google translate, to write your letters.
MOST IMPORTANT STEP
Repeat. Repeat. Repeat. (Method: Deliberate Practice)
You'll most certainly meet some asshole who tells his nonchalant story about mastering Machine Learning in a month, right after you just told your 3-year Data Science journey. Give the guy a cookie, and call Alexa--who responds “I was born knowing Machine Learning”. The point is filter out the noises, because a few will really make you doubt, but I recommend reading what you wrote and understand that you are curious to learn and don't let other discourage you.