This is kind of like putting the cart before the horse. What is "okay" can vary from one environment to the next. Are you experiencing performance problems? If not, then the values are evidently okay. If you are, then before you ask about these specific counters, you should try to determine the nature of the performance problems.
Take temp tables creation rate, for example. Without any other information, there isn't a way to determine whether that is an acceptable value. But if you started to see a lot of contention in tempdb, then you might look at this counter and see if it indicates a lot of tables are being created in tempdb.
Another thing to consider is that what is acceptable for a counter is often determined by historical precedent in your environment. If you haven't already, you should start collecting counters on a regular basis. Establish a baseline so that you know what are typical patterns for performance; when you start to experience performance degradation, you can review counters to see if any are deviating from the baseline.
There are a couple of resources that can help you in the future. First, is a tool call PAL that will analyze performance monitor logs and look for potential problems (please note that these are guidelines, not necessarily actual issues): http://pal.codeplex.com/. The second is a whitepaper detailing a methodology called "Waits and Queues" for finding performance bottlenecks: http://download.microsoft.com/download/4/7/a/47a548b9-249e-484c-abd7-29f31282b04d/Performance_Tuning_Waits_Queues.doc; relevant to your question, this whitepaper talks about several performance counters, how to interpret them, how to correlate counters to identify possible problems, and what are generally good characteristics for the counters.
I hope this helps.