Reading Scientific Journals

I would really like to improve my reading ability with respect to scholarly articles. I have a huge problem of forgetting what I have read after I have read one of these (and that is going to come back and bite me). I currently take notes and annotate my own thoughts on a separate piece of paper (by a separate piece of paper I mean a table in Word). Are these effective methods? what do you guys do?

• Read the first two and last two paragraphs twice to get a feel for what it's going to talk about.
  
• Skim the passage. Look out for any important key words. Get a mental map of the article. Write where I found key words if I have to.
  
• Read the first and last sentence of each paragraph if skimming didn't feel enough.
  
• Finally, read everything word for word at snail pace.
  

So far, it's worked out fine. Reading everything in one go doesn't work for me.

What are the authors trying to do?
What are the authors do to demonstrate their point?
What are the conclusions?

Then branch out from there, like... are their methods sound? Are their assumptions justified?

In short: what is their logic?

edit: it doesn't help that bad papers have poor flow of ideas. What you really need is a good paper with clarity of thought, so that you don't try to remember what they did, but rather their words leave an impression on you. You should be able to sort of second-guess their actions and thoughts, because you need to know what they're saying and not saying, then determine what they need to do in order to prove blah blah blah.

So, you need to have a strong scientific (thought) process inside you first.

edit2: nevermind... while you can read the struck out parts, I might as well give you a more full-fledged thought process.

1) Read the damn abstract. If it doesn't tell you anything, then something is wrong, and either you or the abstract could be in the wrong.
2) What did the authors do/claim/hypothesize/whatever? Any tasty background story?
3) What assumptions did they make, and why? (inside you, you should be constantly evaluating these!)
4) What experiments did they do? Why did they do those experiments? What does each experiment tell you? Do they REALLY tell you those stuff, or are there holes in the stuff the experiments present? Are their analyses valid (based on their assumptions etc.)?
5) What conclusions did they make? Are those conclusions sound based on the data and the assumptions? (What conclusion would you have made?)
6) What future works do they plan? What future works do you think they should do?

Some papers follow this process well - papers on biology and related subjects tend to follow this tightly. Papers on synthesis deviate a bit... but not by much.

Hadriel