The thing labeled "pitch" is actually fundamental frequency (F0), so what Praat displays is an approximation of the rate at which the vocal folds vibrate in producing a certain sound. "Pitch" is a psychological (perceptual) construct which is related but different: the most salient difference is that two signals separated by a fixed amount in terms of F0 will appear to be closer together when the frequency range increases (thus signals of 100 Hz and 120 Hz will be clearly distinct and signals of 10000 and 10020 may not be distinguishable). "Tone" refers to the grammaticalized / lexicalized use of certain larynx-related acoustic properties in some language, such as Yoruba, Japanese, Navaho, Somali, Vietnamese and Chinese: tone is a "package deal", where duration, amplitude, fundamental frequency (including pattern of rise and fall) and voice quality can all be involved: generally, syllables are specified for having some tone. Finally, "intonation" is a similar package deal of frequency / amplitude / phonation etc. modifications imposed on longer strings (clauses and utterances, not syllables), and can coexist with a lexico-grammatical tone system or, as in English and Italian, can combine with word stress.
The "intensity" listing in Praat is the RMS amplitude of the signal, which can be related to (but is not the same as) the perceptual construct "loudness". Intensity does not have a lexico-grammatical correlate, since unlike pitch differences, languages do not employ loudness as a distinctive property of individual sounds. Generally, intensity is a correlate of something else, for example the vowels [a] and [u] differ in intensity because the former has a more open vocal tract and acoustic energy is less suppressed and the latter has a more closed vocal tract and more suppression of acoustic energy. At the level of intonation (over longer spans such as a whole word), you can shout a word and that will increase the intensity of the segments therein.
"Stress" is a package deal similar to "tone", but with a vastly reduced set of possible distinctions. Whereas languages with tone may have a dozen different tones, a language with stress has at most primary and secondary stress, plus unstressed. Unlike tone systems where you generally have to learn for each syllable what the tone is, in most languages with stress, there are rules that say which syllable has the stress (and most often the distinction is just between stressed and unstresed). English is a bit extreme in that respect, since it is very hard (ultimately, impossible using reasonable rules) to fully predict where stresses are, and which one is more prominent (for example the words tigress, the verb progress and the noun progress exemplify 1-0, 0-1 and 1-2 stress patterns. Generally speaking, stress translates into "increased pitch, amplitude and duration".
Finally, "pitch accent" refers to languages which are somewhere between "tone" and "stress", and can refer to quite a range of languages (this indeterminacy is why phonologists have generally rejected the concept as a valid category). The main characteristic of a so-called pitch-accent language is that is has limited tonal contrasts within a word, so if there are two "accents" in a language, a 4-syllable word allows fewer than 16 contrasts (which you can get in a two-tone system). A secondary characteristic of pitch accent systems is that there can be phonetic distinctions such as "rising" versus "level" on a syllable, whereas with plain stress systems, if a syllable is stressed, it doesn't then contrast "stressed and rising" versus "stressed and level" (if it does, people tend to call is a pitch accent system).
Praat is about the acoustic properties of the signal, and not about the phonological analysis. It cannot give you a phonological analysis, but it will tell you what the phonetic properties of an utterance are, which you then have to translate into a phonological system.
