Here is how I always look at it:
The items on the right should serve the items on the left.
Individuals and Interactions over Processes and Tools
We have all known processes or tools that help us do our job better. We've also all known processes or tools that may have once served a purpose, but now just get in the way. As long as the process or tool helps individuals with their interactions with others, they're good. When they hinder interactions, we should look at more effective processes and tools.
Working Software over Comprehensive Documentation
There is documentation that helps us build great software. There is documentation that helps us maintain the software after it is done. Then there is documentation that sits in a drawer, never to be read. We should write a sufficient amount of the type of documentation that we need, not more.
Customer Collaboration over Contract Negotiation
Many contracts are about protecting parties from each other, not improving how the team and their customers interact. A good contract can form the basis for a great, collaborative interaction between the two parties.
Respond to Change over Following the Plan
Plans change all the time. When we invest too heavily in a plan, we become attached and it feels safe to follow, even if we know we're following it off a cliff. If a plan is built by a committee of smart, high-payed people and it fails, it's their fault, not mine. But if I see it failing and I change, I take responsibility. This is why there are so many examples of massive failures where the people following the plan kept following it long after it was clear it wasn't working.
That doesn't mean planning is bad though. With emergent just-in-time planning approaches, we can gain enough context to make educated guesses about how the future will unfold while at the same time we can be ready to pivot as soon as the need is apparent.