Does Lightsail's Instances quota is the maximum instances you could create, not the maximum instances could run simultaneously? Is the such `quota` meaning still apply in EC2?

I was trying lightsail these days, and found there is a quota limit named "lightsail instance" whose description is "The maximum number of instances per region". But in practice, it is actually the maximum number of instances per region you could create, not the maximum number of instances running simultaneously. In my case, it's 20, which means if I create 20 instances run for some certain period and delete them, afterward I cannot create another single instance anymore, because I already created 20 instances! Is that normal and is it designed on purpose? If it's true. That is a very strict restraint.

My 2nd question is whether such quota meaning still applies in EC2? For example, in the service quota of EC2, there is a quota named "All Standard (A, C, D, H, I, M, R, T, Z) Spot Instance Requests" whose description is "The maximum number of vCPUs for all running or requested Standard (A, C, D, H, I, M, R, T, Z) Spot Instances per Region". In my case, the value is 32. Does the same restraint mechanics like Lightsail instance quota still apply here? If true, this means if I create 32 vCPUs spot instances and terminate them, then I cannot create another single EC2 spot instance anymore. Or is this quota really for the number of running Spot Instances and pending Spot Instance requests per AWS account per Region?