Which computational model has the same language-recognition power as a standard deterministic single-tape Turing machine?
A multi-tape TM can be simulated by a single-tape TM, so both recognize exactly the recursively enumerable languages.
For $\delta (q0, a) = (q1, b, R), what$ is the immediate effect of the transition?
The transition writes b in the current cell, moves right, and changes the state to q1.
What is the standard asymptotic upper bound for simulating a k-tape TM running in T(n) time by a single-tape TM?
The standard single-tape simulation uses O(T(n)^2) time.
Which statement about nondeterministic Turing machines is correct?
Nondeterminism does not increase TM computability power, although it can affect time complexity.
Which model is strictly weaker than a Turing machine in language-recognition power?
A one-stack PDA recognizes exactly the context-free languages, a proper subset of the recursively enumerable languages.