Don't be scared that your skimming over topics - general knowledge is good - just make sure that occasionally you do a deep dive into something - spend some time looking in detail at a ML topic or language feature etc.
A deep dive doesn't need to be long - committing to 15 minutes learning about python lambda functions is time well spent. Those 15 minutes will add up and if you enjoy the topic will inspire you to do more.
Create structured notes on topics, if you read an ML book chapter make a note of the important bits, watch a youtube video write down any things that stuck out with a link, write some code fragments that demonstrate an API.
No one book can teach you a subject (I found this particularly true with ML), If you find yourself loosing interest reading a book/chapter, don't carry on - your probably not learning anything. Find some other resources on the topic and come back to it later (or not)