I agree with you that five years might as well be a lifetime. The point of this exercise is to define how you want that lifetime to end, then step backwards through it until you know what you're doing tomorrow. The plan for five years ("be a CTO") only matters insofar as it tells you your plan in three years ("be in a position where you report to the board"), one year ("be a lead engineer"), one month ("be confident in passing a job interview and be sending my CV out"), and tomorrow ("message Todd and ask if he'll run a mock interview for me, do some leetcode, message the Acme group chat").
You honestly might as well throw out any plans beyond the one year mark. Either they're important and you can recreate them, or they've changed and you should recreate them. The process of planning is more important than its output.
I suppose this might work for some, but it comes off as excessively performative and not actually practical.