The title "Project Control Cycle" gives a clue: here's how I would read it.
The point of the diagram is that it is a feedback cycle for project control. Let's start in the middle: in order to make project decisions, you need objectives (what you're trying to accomplish) and information (how you're doing so far, what your users want, what the competition is like, etc etc).
Defining what you're trying to accomplish is pretty straightforward, but you can't get information from the world - you can only get data. You collect raw data from the real world (eg, keyclicks), and then you probably process it iteratively (eg, turn it into analytics). Now it's useful information.
In order to decide what to do in order to accomplish your goal given that information, you probably have to do some modelling, whether formal or informal. EG, "People are only spending 2 min on my website, and I want them to spend 5 min. What will probably happen if I add a rotating gallery? What about an article with a comment box? What if I rearrange what's there?"
That modelling process will eventually lead you to a decision: in your models, the article with a comment box was most likely to get people to spend 5 min there. So now you have an informed decision and/or plan: "We will add an article with comment box to the website. This will require (assorted task breakdown and schedule)."
The "Implementation" stage is "implement the plan". Do the thing you decided to do.
And now... it's done... it's deployed out in the real world.. and you start gathering data from the real world again. Are people now spending 5 min on your website? What else has changed? What's your next objective? And so forth. Lather, rinse, repeat.